Glycoconjugate Journal

https://doi.org/10.1007/s10719-018-9812-0

ORIGINAL ARTICLE

Differential recognition of natural and remodeled glycotopes

by three Diocleae lectins
Tania M. Cortázar 1 & Iain B. H. Wilson 2 & Alba Hykollari 2 & Edgar A. Reyes 1 & Nohora A. Vega 1

Received: 12 September 2017 / Revised: 5 January 2018 / Accepted: 10 January 2018

# Springer Science+Business Media, LLC, part of Springer Nature 2018

Abstract
The carbohydrate specificities of Dioclea grandiflora lectins DGL-I1 and DGL-II, and Galactia lindenii lectin II (GLL-II) were
explored by use of remodeled glycoproteins as well as by the lectin hemagglutinating activity against erythrocytes from various
species with different glycomic profiles. The three lectins exhibited differences in glycan binding specificity but also showed
overlapping recognition of some glycotopes (i.e. Tα glycotope for the three lectins; IIβ glycotope for DGL-II and GLL-II lectins);
in many cases the interaction with distinct glycotopes was influenced by the structural context, i.e., by the neighbouring sugar
residues. Our data complement and expand the existing knowledge about the binding specificity of these three Diocleae lectins, and
taken together with results of previous studies, allow us to suggest a functional map of the carbohydrate recognition which illustrate
the impact of modification of basic glycotopes enhancing, permiting, or inhibiting their recognition by each lectin.

Keywords Diocleae . Dioclea grandiflora . Galactia lindenii . Lectin specifity . Remodeled glycoproteins . Hemagglutination

Introduction integral cell membrane glycoconjugates are of great impor-

tance in many normal and pathological biochemical process-
The specific, reversible and non-covalent interactions between es; due to their many roles, including ones in cell communi-
lectins and their carbohydrate ligands attached to soluble and cation, tumor metastasis, innate immunity, sperm-egg recog-
nition, infections by parasites and, plant protection against
phytopathogents, lectins used for many analytical, diagnostic
Electronic supplementary material The online version of this article or therapeutic applications [1, 2]. Lectins occur in all king-
(https://doi.org/10.1007/s10719-018-9812-0) contains supplementary
material, which is available to authorized users. doms of life, but historically important are those from plants,
especially from legumes. Due to their high abundance in
* Tania M. Cortázar beans, there have been many studies on bean lectins, including
[email protected] concanavalin A (ConA) isolated from Canavalia ensiformis, a
member of the Diocleae tribe. Recently, a comprehensive mo-
Iain B. H. Wilson lecular phylogenetic analysis based on nuclear and chloroplast
[email protected]
markers and supported by morphological features allowed the
Alba Hykollari recognition of the three major lineages within the tribe
[email protected] Diocleae of papilionoid legumes: Canavalia, Dioclea and
Edgar A. Reyes Galactia clades [3]. In the present study, we examined the
[email protected] binding specificities of three Diocleae lectins: DGL-I and
Nohora A. Vega DGL-II lectins found in the Brazilian native Dioclea
[email protected] grandiflora Mart ex. Benth species [4, 5]; and GLL-II, one
of the lectins present in Galactia lindenii Burkart [6], plant
1
Protein Research Group, Department of Chemistry, Universidad which grows endemically in Colombia and Perú [7].
Nacional, Calle 45 # 30-03, Building 451. Lab. 201-1,
DGL-I has been characterized before by others and found
Bogotá, Colombia
2
to display carbohydrate specificity towards some mannose-
Molecular Glycobiology Research Group, Department für Chemie,
and glucose- (Man/Glc) containing carbohydrates [4, 8–13];
Universität für Bodenkultur (BOKU), Muthgasse 18,
A-1190 Wien, Austria it also has the ability to agglutinate rabbit RBCs [4, 8, 10, 12];
 Glycoconj J

additionally, the sequence of its subunits have already deter- 8 h each, in PBS containing 5 mM thiourea, pH 7.2, at 4 °C
mined [14, 15]. On the other hand, previous studies in our with gentle agitation. The three extracts were pooled and cen-
laboratory have shown DGL-II [5] and GLL-II [6] to share trifuged (15,000 x g, 30 min, 4 °C); the supernatants were
some characteristics such as: their N-terminal amino acid re- dialysed (3×) against 20 mM NH 4HCO 3 and stored at
gions present a unique sequence hitherto found only in some −20 °C until required. Subsequently, the dialysates were first
Diocleae lectins (designated type II); both lectins recognize subject to ion exchange chromatography. Then, the fractions
the trisaccharide representing H-type II blood group glycotope obtained were subjected to affinity chromatographies using
[α-L-Fuc(1-2)-β-D-Gal(1-4)-β-D-GlcNAc-O-R] and aggluti- different conditions for each lectin. Briefly, clear supernatants
nate some types of red blood cells (RBCs) but not those from were applied (~15 mL) to a column packed with DEAE-
rabbit. But also, specifically interact with lactose (Lac)-sepha- Sephadex A-50 equilibrated in PBS. Subsequently, the un-
rose supports, thus showing they are different from the Man/ bound material (DEAE peak I) was fractionated in
Glc-specific lectins. Sephacryl S-200 and the DGL-I lectin was desorbed from
The present study reports the binding specificity screening column with 0.2 M glucose in PBS. On the other hand, the
of DGL-I, DGL-II and GLL-II lectins towards 11 different DEAE retained fraction was eluted using PBS containing 1 M
glycoconjugates, enzymatically remodeled from human apo- NaCl, and later applied to a Sepharose-Lac support previously
transferrin and bovine fetuin to produce proteins with defined prepared by coupling ethanol-washed lactose to divinyl
glycan structures [16]. The results of Diocleae lectin- sulphone-activated Sepharose 4B [17]. DGL-II lectin was
glycoconjugate interaction were combined with data of their eluted with PBS-0.2 M lactose. Changes of buffer and protein
hemagglutinating activity towards RBCs from different spe- concentration were performed by ultrafiltration using cellu-
cies which express distinct sets of glycoantigens, allowing us lose membranes (NMWL: 1 kDa for DGL-I and 10 kDa for
to complement and expand the knowledge about the specific- DGL-II). Eluted fractions were monitored by absorbance at
ity of these three Diocleae lectins as well as propose a map of 280 nm. Purification of GLL-II lectin was carried out as de-
glycotope specificity for each of them. scribed [6]. Protein was determined by the bicinchoninic acid
(BCA) assay [18]. SDS-PAGE were performed according to
Schägger & von Jagow [19] for DGL-I and to Laemmli [20]
Materials and methods for the lectins type-II. MALDI-TOF MS/MS of two tryptic
peptides (m/z 1346 and 3255) from DGL-I, as compared to
Materials the published sequence, verified the identity the isolated lec-
tin, whereas the peptide map of DGL-II was distinct but
D. grandiflora seeds collected in Carto, Ceará (Brazil) were matches no protein sequence in the databases; distinct staining
kindly provided by Dr. R. Moreira (University of Fortaleza, upon immunoblotting with anti-horseradish peroxidase was
Brazil). Galactia lindenii seeds collected in Cundinamarca compatible with the presence of a typical plant N-glycan on
(Colombia) were botanically identified at the Natural DGL-II.
Sciences Institute (ICN) from National University of
Colombia (voucher COL 15115). Pharmacia and Bio-Rad Dot blots
e q u i p m e n t w a s u s e d f o r c h r o m a t o g r a p h y.
Biotinamidohexanoic acid 3-sulfo-N-hydroxysuccinimide es- Dot blot assays were performed following the method de-
ter sodium salt, protein standards, neuraminidase from scribed [16] and the biotinylation of the purified lectins
Clostridium perfringens and β1,3-galactosidase were obtain- allowing the coupling to primary amine residues [21].
ed from Sigma–Aldrich–Fluka (St. Louis, MO, USA). Glycoproteins (~ 20 mg/mL) were diluted (1:10) and 0.5 μl
Alkaline phosphatase-conjugated streptavidin were purchased of these dilutions was spotted onto nitrocellulose (BioTrace
from Vector Laboratories (Burlingame, CA, USA). Fresh hu- NT, Pall Life Sciences, East Hills, NY, USA). The samples
man blood was obtained from the Clinical Laboratory and were left at RT to dry completely, and subsequently the mem-
animal erythrocytes were supplied by the Veterinary branes were blocked overnight at 4 °C in TTBS (Tris-buffered
Faculty’s Haematology Laboratory of the National saline containing 0.05% Tween; i.e., 0.1 M Tris-HCl, 0.1 M
University of Colombia. Sugars were commercial products NaCl, pH 7.4, 0.05% Tween) supplemented with 0.5% bovine
having the highest available purity and the rest of reagents serum albumin (BSA). After another washing step, mem-
were of analytical grade. branes were incubated for 1 h on a rocking table at RT with
the biotinylated lectins diluted in TTBS with 0.5% BSA to a
Extraction and purification of lectins concentration of 5 μg/mL. Subsequently, membranes were
washed again three times prior to incubation with the alkaline
Soluble proteins were extracted from D. grandiflora seed phosphatase streptavidin conjugate (1:1000). After incubating
defatted flour (1:10, w/v) by three consecutive extractions of for 1 h at RT under shaking, the membranes were washed
Glycoconj J

Fig. 1 Purification of Dioclea grandiflora lectins. a Fractions from„

DEAE column. b Affinity chromatography for DGL-I. c Affinity
chromatography for DGL-II. d SDS-PAGE staining with Coomassie
blue. E1: 1st extraction; E2: 2nd extraction; ET: total saline extract; I:
DEAE-unretained fraction; II: DEAE- retained fraction. DGL-I and
DGL-II: purified lectins. Absorbance reading at 280 nm

another three times and then finally incubated in the develop-

ing solution (5-bromo-4-chloro-3-indolylphosphate/nitro blue
tetrazolium, Sigma-FAST tablets).

Hemagglutination

Lectin hemagglutinating activity was assayed by serial dilu-

tions on microtiter plates using 2% (v/v) RBC suspensions
[22]. RBCs exposing T or Tn determinants were prepared
from suspensions of A erythrocytes subjected to enzymatic
treatment using α-sialidase and β-galactosidase [23]. The spe-
cific titer was defined as the agglutination titer divided by the
protein (μg) in the assayed solution. The minimum aggluti-
nating concentration represents the lowest tested concentra-
tion in a serial dilution at which cell agglutination was visible.

Results

Lectin extraction and purification

Dioclea grandiflora lectins were obtained by seed flour salt

extraction, separated in anion exchange support, and purified
by affinity chromatographies. For extraction, 5 mM thiourea
was included in PBS pH 7.2, to reduce the amount of poly-
phenols in the seed extract by inhibiting polyphenol oxidases,
as previously used when working with pigmented extracts
from Diocleae and Lamiaceae species [6, 24]. The concentra-
tion of soluble protein in total saline extract was 4.4 mg/mL
and this extract strongly agglutinated all RBCs types assayed
in this study (A, B, O, T, Tn, pig and sheep RBCs). The
protein fractions obtained in each step of purification were
used for protein, agglutination and SDS-PAGE assays, and
in the subsequent steps. The specific activities were differen-
tiating according to the lectin and RBC type through the pu-
rification steps.
D. grandiflora saline extract was initially fractioned on
DEAE-Sephadex A-50 support equilibrated in PBS
(Fig. 1a). DEAE exchange chromatography permitted us to
separate two lectins present in the same seed flour taking into
account their pI, insomuch as Moreira et al. showed DGL-I is
a basic protein with components of isoelectric points (pI) be-
tween 8.6 and 9.0 [4], and in our laboratory it was found DGL-
II is slightly acidic protein with pI between 5.1 and 5.4 [5].
Both DEAE- unretained (DEAE URF) and retained (DEAE
RF) fractions exhibited agglutinating activity (Table 1). Then,
 Glycoconj J

3.68

NA
ND
ND
1.0
both fractions were subjected to separate affinity chromatogra-

S
phy steps. The DEAE URF was fractionated on the crosslinked

7.37

2.38
ND
ND
1.0
P
dextran copolymer support Sephacryl S-200 and the DGL-I

7.37
0.44

NA
NA
1.0
Tn
lectin was recovered by elution with 0.2 M glucose in PBS

38.05
11.90
(Fig. 1b). SDS-PAGE revealed that the affinity chromatography

7.37
3.85
1.0
T
Purification (fold)
resulted in removal of other bands of higher molecular weight

1.84
0.44

NA
NA
observed in DEAE URF (Fig. 1b, d); purified DGL-I showed

1.0
O
the typical pattern observed before for this [4] and other

3.68
0.88

NA
NA
1.0
B
Diocleae type I-lectins [25–41], as it is composed of one intact

3.68
0.88

NA
NA
subunit (α; 27.3 kDa) and two fragments (β and γ; 14.7 kDa

1.0
A

DEAE URF, DEAE-unretained fraction; DEAE RF, DEAE-retained fraction; NA, non-agglutination; ND, undetermined; P and S, pig and sheep RBCs, respectively
and 12.1 kDa) originating from post-translational processing of

34.07
9.24
ND
ND
ND

NA
ND
ND
the lectin precursor during seed development, corresponding to

68.14

21.97
the originally covalently-bound N- and C-termini [42]. In the

9.24
ND
ND
ND

ND
ND
P
present study, SDS-PAGE with tricine [19] facilitated a better

68.14
4.07
9.24
resolution of the small subunit fragments. On the other hand,

ND
ND
ND

NA
NA
Tn
DEAE RF was applied onto a Sepharose-Lac column after three

68.14

351.6
35.58
9.24
ND
ND

110
ND
PBS buffer changes in order to remove salt as well as remaining

T
pigment. DGL-II was eluted with PBS-0.2 M lactose (Fig. 1c).

17.03
2.92
0.86
5.82

9.24
4.07

NA
NA
DGL-II subunit molecular weight determined by Laemmli

O
Specific titer a

34.07
SDS-PAGE was 25.4 kDa (Fig. 1d). The data obtained during

2.92
1.72
5.82

9.24
8.15

NA
NA
B
the extraction and purification process are shown in Table 1.

34.07
1.46
0.86
5.82

9.24
8.15

NA
NA
A
Glycoprotein binding

1:1024
1:2048
ND
ND
ND

NA
ND

ND
S
We used native and remodeled glycoproteins with multiple

1:2048

1:2048

1:128
glycosylation sites to delve deeper in the specificities of the

ND
ND
ND

ND

ND
P
DGL-I, DGL-II and GLL-II lectins, considering that in nature,

Specific titer defined as agglutination titer divided by protein (μg) in the assay well (50 μL lectin)
1:2048
1:2048
1:128
the lectin-carbohydrate interactions are combinations of sever-
ND
ND
ND

NA
NA
Tn

al simultaneous binding events carried out with multivalent

1:2048
1:1024
1:2048
1:2048
oligosaccharide ligands, instead than a simple individual bind- 1:2048
ND
ND
ND
Vol (mL) Protein Total protein (mg) Erythroagglutinating titer

ing event [16]. They were prepared various defined glycoforms

Affinity fractions on T, Tn, P and S erythrocytes were assayed in 25 μL lectin/well

1:2048

1:2048

1:1024
of two glycoproteins: human transferrin, which carries two
1:512
1:128

1:256

NA
NA
O

biantennary complex-type N-linked carbohydrate chains with-

1:2048

1:2048

1:2048

in its C-terminal domain [43], and bovine fetuin, the major

1:512
1:256

1:512
Purification of D. grandiflora seed lectins DGL-I and DGL-II

NA
NA

glycoprotein in fetal calf serum with three O-linked glycans

B

1:2048

1:2048

1:2048

[44] and three N-linked glycans [45]. Remodeling of glycopro-

1:256
1:128

1:512

NA
NA

teins was performed as described (Supplemental Fig. S1). To

A

prepare the tranferrin neoglycoforms, the native protein was

desialylated and degalactosylated prior to further modification
with recombinant glycosyltransferases as before [16].
161.184
124.656
260.554

Structures in the remodeled forms of asialoagalactofetuin

553.87

13.41
60.33
15.14

3.08

(F2) and asialofetuin (F3 and F4) result from various combi-
nations of sialidase and galactosidase treatments. F3 was also
(μg/ mL)

treated with PNGaseF and, therefore, have no N-glycans. After

3504
2968
7042

4431
1257
1202
745
233

enzymatic remodeling, the resulting neoglycoforms display

defined carbohydrate sequences were spotted on nitrocellulose
(~ 1 μg) to identify the sugar moieties bound by the lectins
13.2
12.6
125

18
46
42
37

48

(Fig. 2). Also, a sample containing naturally-biotinylated pro-

Pool saline extracts

teins was used as a positive control for streptavidin binding.

Purification step

In the dot blot, we could see DGL-I bound 8 of the 11

DEAE URF

Lac-affinity
Glc-affinity
DEAE RF

glycoprotein forms tested, half of them with N-glycans in

Extract 2
Extract 3
Extract 1
Table 1

their structure, and the other half with both N- and O- gly-
cans (Fig. 2 and Supplemental Fig. S1). On the other hand,
a
Glycoconj J

Fig. 2 Screening for ligands of galactose-, N-acetylgalactosamine-, sialic GnGnF6, i.e., core α1-6-fucosylated asialoagalactotransferrin. Fetuin
acid and fucose- binding Diocleae tribe lectins. DGL-I: D. grandiflora glycoforms: F2: asialoagalactofetuin with N- and O-glycans; F3: asialofetuin
lectin I; DGL-II: D. grandiflora lectin II; GLL-II: G. lindenii lectin II. with only O-glycans; F4: asialofetuin with N- and O-glycans and F5: native
Transferrin glycoforms: Sia2,6: native glycoform with two terminal fetuin with N- and O-glycans. BC: biotinylated protein control. The right
NeuAα2-6 residues; Gal: GalGal (asialotransferrin); Gn: GnGn hand panel shows the respective example structures according to the
(asialoagalactotransferrin); βGn: remodelled glycoform with terminal Standard Nomenclature for Glycans (for more details see also
βGalNAc residues; αGal: remodelled glycoform with terminal α-galactose Supplementary Fig. S1)
residues; MM: pentasaccharide core with terminal mannose residues; GnF6:

DGL-II and GLL-II showed more specific behavior Agglutination of erythrocytes with different glycomic
interacted with only 2 or 4 glycoprotein forms, respectively; profile
both type-II lectins showed preference for O-glycans as they
did not bind to any transferrin glycoforms (Fig. 2). In parallel to the dot blot analyses, we carried out hem-
Specifically, we could observe DGL-I and GLL-II recog- agglutination activity assays evaluating serial dilutions of
nized asialofetuin (F3 and F4, i.e., forms with and without lectins on RBCs of different species. As histo-blood
N-glycans) but not asialotransferrin (β1-4-Gal linked) (Fig. 2; groups are non-reducing terminal elements of the oligo-
and Supplemental Fig. S1), thereby indicating binding to O- saccharide chains in glycosphingolipids and glycoproteins
glycans of the form Galβ1-3GalNAcα and/or Galβ1- expressed on RBC and cells [46], this approach allowed
4GlcNAc (glycotopes Tα and/or IIβ). DGL-I also bound us to test multivalent oligosaccharide ligands in the con-
asialoagalactofetuin (F2, which contain O-glycans trimmed text of a cellular surface environment [47].
to the GalNAcα1-O-Ser/Thr or Tn antigen core) as well as Based on various data, it can be estimated that one human
asialoagalactotransferrin form βGnβGn containing β1-4- RBC may have up to 8 × 105 antigen sites in total on the
linked GalNAc termini, suggesting a preference of this lectin AB(H)-carrying glycolipids of its surface [48], in addition to
for terminal HexNAc regardless of the linkage or glycan type a small number of abundant histo-blood group antigen carrying
(O/N). GLL-II also bound α-linked GalNAc (as found on F2) proteins [49] (Supplemental Table S1). On the other hand, in-
although to lesser extent. On the other hand, DGL-II complete synthesis of carbohydrate chains results in the pres-
interacted very weakly with F2 and not at all with ence of tumor associated Tα and Tn glycotopes on the cell
asialoagalactotranferrin (GnGn) nor with any of the other surface [50]. The highly glycosylated (60%) transmembrane
forms of N-glycans. DGL-I, DGL-II and GLL-II (although glycophorin A (GYPA) containing ca. 12 O-glycans (especially
weaker) bound native fetuin (F5) but not native transferrin NeuAcα2-3Galβ1-3(NeuAcα2-6)GalNAc) and one N-glycan,
suggesting that α2-3-linked sialic acid (as found on fetuin is the major carrier of sialic acid-containing glycans [51]. Thus,
but not on transferrin) could be a ligand of these Diocleae A-type erythrocytes were treated with sialidase thereby yielding
lectins. While the presence of core α1-6-fucose had little ef- RBCs that contain asialo-GYPA exhibiting the Tα glycotope;
fect on binding of any of the three lectins, none of the lectins another group of A-type RBCs was sequentially treated with
interacted with either asialotransferrin (GalGal) or its α1-3- sialidase and β-galactosidase resulting in erythrocytes contain-
galactose capped form (αGalαGal); thus Gal is not an obvious ing asialoagalacto-GYPA with exposed Tn glycotope [23, 52].
ligand of any of these three Diocleae tribe lectins. Antigenic determinants in glycans present in the cellular
 Glycoconj J

surface of different types of RBCs, including histo-blood anti- Table 2 Minimum agglutinating concentration of Diocleae Tribe
lectins (μg/mL)
gens from human [46, 53] as well as from animal cells [53–57]
are indicated in the Supplemental Table S2. Erythrocyte (RBC) type
The D. grandiflora seed saline extract showed strong ac-
tivity towards all RBC types assayed in this study, with ag- Lectin A B O T Tn P S
glutination even at the highest dilution evaluated (Table 1). DGL-I 0.58 0.58 1.17 0.58 0.58 0.58 1.17
DEAE URF fraction showed activity on all RBC suspensions
DGL-II NA NA NA 0.11 NA 1.82 NA
to a greater or lesser extent depending on erythrocyte type. In
GLL-IIa 28 7.0 7.0 ND ND ND NA
the same way, affinity purified DGL-I agglutinated all type of
RBCs assayed exhibiting differential specific titers, i.e., this NA non-agglutination, ND undetermined
lectin strongly agglutinated 2% pig, T and Tn RBC suspen- a
Data from Ref. [6]
sions with approx. 15 ng of lectin in the last agglutinating
dilution assay well tested, while it was needed 60 ng for O from highest to lowest degree of agglutination by DGL-I the
erythrocytes (Table 1). RBCs presenting the A and B histo- following order of RBC type is obtained: horse > rabbit, dog
blood groups as well as the sheep RBCs were agglutinated to > pig, T, Tn > A, B ≥ sheep > O.
a similar extent by DGL-I with a specific titer 2 fold greater On the other hand, DEAE RF fraction was able to aggluti-
than those for O-type. Our results agree to that observed for nate T-type RBCs as well as purified DGL-II showed a strong
some Dioclea spp. type I (BMan/Glc-specific^) lectins, which specific activity towards T-erythrocytes, with a specific titre
are able to agglutinate both human and animal erythrocytes almost 5 times higher than that given by the type I lectin on the
[36–41]. Moreira et al. showed the agglutinating activity of same RBC type (Table 1). The minimal agglutinating concen-
DGL-I on rabbit RBCs was dropped to 1.4%, 3.8% or 32.4% tration tested of DGL-II for T-type RBCs was 0.1 μg/mL
in presence of Man, fructose (Fru) or Glc, respectively, in the (2.8 ng of lectin/last well); while its activity on pig erythrocyte
same concentration (100 mM); while the presence of galac- was weaker and the rest of RBCs type suspensions used in the
tose (Gal) had no effect at all [4]. Furthermore, it was dem- present study were not agglutinated by DGL-II (Table 2). This
onstrated DGL-I activity on rabbit RBCs also is strongly result differs from that found for other Diocleae tribe type II-
inhibited by the trisaccharide 3,6-di-O-(α-D- lectins, which did agglutinate normal human RBCs [6,
mannopyranosyl)-D-mannose present in the core region 37–39], as the case of GLL-II whose minimal concentration
(e.g., MM in Supplemental Fig. S1) of the asparagine- required was 28 μg/mL for A, and 7 μg/mL for B and O
linked carbohydrates and by the Man9-Asn glycopeptide, erythrocytes [6]; Dioclea lehmanii lectin II (DLL-II) which
both at micromolar concentrations [12]. This lectin is able agglutinated RBCs presenting those three histo-blood groups
to agglutinate other animal RBCs [38] and the agglutination to a similar extent requiring 0.2–0.3 μg of lectin [38]; and
of normal human erythrocytes had already been observed [11, Dioclea sericea lectin II (DSL-II) which also agglutinated
38]. DGL-I agglutinated dog and horse RBCs with specific these three RBC types showing a higher titre for type B eryth-
titers 2.5 and 5 times greater than that of normal human rocytes compared to that obtained with erythrocytes A and O
RBCs, respectively; besides rabbit RBCs were agglutinated [39]. The Diocleae type II-lectin preference towards different
by the lectin with the same potency as those of dog, while in animal RBCs also varies; e.g., GLL-II showed a marked dif-
accordance with our results its activity towards sheep RBCs ference in the specific titers for rat and rabbit RBCs, requiring
was weaker [38]. It had also been observed similarities and 3.7- and 50-fold, (respectively), the concentrations needed to
differences in the concentration of the same sugar depending human O erythrocytes, and no agglutination was detected with
on the type of RBC in agglutination inhibition assays for mouse, cow, horse, dog, and sheep RBCs using up to 1.4 mg/
DGL-I lectin. Thus, agglutination of O [11] and rabbit [10] mL lectin [6]. DLL-II agglutinated sheep RBCs [38]; mean-
erythrocytes was inhibited by a minimal concentration of while DGL-II did not have activity on rabbit nor sheep RBCs
2.1 mM of α-D(+)Man and 16.7 mM of α-D(+)Glc for both [5]. Similarly, DLL-II strongly agglutinates horse RBCs
RBC types, while trimannoside at 12 μM or 30 μM was meanwhile GLL-II does not at all [6, 38]. So, GLL-II agglu-
required to inhibit rabbit- [12] or O [11] RBC agglutination, tinated B/O, rat and A RBCs, in that order, from highest to
respectively; and GlcNAc inhibited O RBC at 16.7 mM [11] lowest specific title [6], meanwhile for DGL-II was obtained:
while not at all rabbit RBC agglutination, even at 100 mM T > > pig (Table 1.). This is to our knowledge, the first report
[12]. About activity on T/Tn RBCs, we observed DGL-I ag- regarding either agglutinating activity of D. grandiflora
glutinated both types in the same extent (Table 2), with a lectins towards pig-RBCs, DGL-I activity on T- and Tn-
minimal tested concentration 3.4 times greater than the min- RBCs or specific activity of DGL-II towards T- RBCs.
imum concentration necessary of the Tn-specific Salvia Besides, through binding assays to neoglycoconjugates,
bogotensis lectin [24]. Taking into account the specific titers previously in our laboratory it was found that DGL-II and
obtained in the present study and those from [38], in sequence GLL-II specifically recognize the H-type II blood group,
Glycoconj J

while binding to H-type I, H-type III, H-type IV, Lea or Ley (aSBM) coupled to Sepharose 4B was unable to retain the
determinants was weaker [5, 6]. Amongst the tested carbohy- GLL-II lectin. However, in keeping with the abundance of O-
drates, DGL-II was only inhibited by di/trisaccharides having linked GalNAc on the protein, ELLSA assays showed the
no apparent structural relationship (e.g. 25 mM melezitose, GLL-II bound to aSBM linked to the polystyrene plate [6].
40 mM saccharose or 90 mM lactose) [5]; while GLL-II was Also, the activity of this lectin was inhibited but weakly (mM
inhibited by 12.5 mM N-acetyl-D-galactosamine (GalNAc), order) by the free monosaccharide GalNAc [6]. In the present
50 mM lactose or 50 mM Me-α-D-galactopyranoside, but study, we observed there was diminished or absent GLL-II
did not by melezitose or saccharose [6]. GLL-II activity also interaction with the terminal GalNAc(α/β)-containing struc-
was inhibited by secretor Leb salivas (1:100 dilution; [6]). tures, thus we conclude that GLL-II binding to GalNAc is often
Both type II-lectins were not inhibited by D-Man or D-Glc. low regardless of whether this sugar is on O/N-glycans or on
The relatively high sugar concentrations of mono-, di- and glycolipids.
trisaccharides needed to inhibit erythroagglutination indicated
DGL-II and GLL-II probably could recognize more complex
structures, including possibly some containing fucose. Galactose-containing glycotopes

Although Gal as a monosaccharide is not a good ligand

Discussion for either DGL-I or GLL-II [4, 6, 10, 12], consistent
with previous data, the lectin DGL-I was capable of
Lectin specificity towards different glycotopes binding to structures containing Galβ1-3GalNAc, as in
the Tα glycotope-presenting asialofetuin forms F3 and
In the present study, we examined three Diocleae lectins F4, and T-type RBCs. GLL-II also showed specificity
(DGL-I, DGL-II and GLL-II) and their binding to mammalian towards exposed Tα glycotope interacting with F3 and
glycoproteins and to RBCs with different glycomic profiles; F4 glycoforms, while DGL-II did not show interaction
while the glycoproteins have defined glycosylation profiles, with F3 or F4, although its activity on T-RBCs obtained
agglutinination tests interactions in ‘real’ cell membrane envi- the higher specific titer in the present study, which may
ronments. Thus, different contexts and densities of glycotopes be due to the different modes of Tα glycotope presen-
were investigated. In the following paragraphs, we attempt to tation between a glycoprotein and a cell surface. It has
explain the specificities of DGL-I, DGL-II and GLL-II been reported for some lectins the method of analysis
refering to data from the present and selected previous studies may hinder the identification of differential specificity,
to shed light on their fine specificities. depending not only on the glycan structure but whether
it is surface bound or in solution, as has been observed
GalNAc glycotopes for mammalian galectin-1 [61].
None of the three lectins showed interactions with
Various HexNAc-based structural units occuring in mammals Gal(α/β)-exposing remodeled N-glycoforms (αGalαGal
are commonly galactosylated: GalNAcα1-Ser/Thr (Tn core) and GalGal) which contain, respectively, the Galili anti-
and Galβ1-3GalNAcα1-glycotope (Tα), both for O-glycan; g en (G a l α 1 - 3G alβ 1 - 4 G l cN A c β ) or e xp os ed N -
Galβ1-3GalNAcβ1-glycotope (Tβ) for glycoconjugates and acetyllactosamine (LacNAc). Thus, while these data are
glycosphingolipids; and, Galβ1-3GlcNAcβ1- (Iβ) and compatible, e.g., with GalGal not inhibiting the activity
Galβ1-4GlcNAcβ1- (IIβ) for O/N-glycans [58–61]. GalNAc- of DGL-I even at 1.6 mM [12], the high specific titer
recognizing lectins often either show (i) broad specificity for α/ obtained for DGL-I in the activity on pig RBCs used in
β-linked or internal GalNAc structures or (ii) only recognise the present study and observed before for rabbit erythro-
terminal αGalNAc with some involvement of the subterminal cytes [38] could be due to the presence of αGal
sugars [16, 24, 60]. We observed DGL-I recognized structures glycotope on glycolipids. However, the cryptic
which exhibited exposed GalNAc regardless of the type of (sialylated) form of IIβ glycotope in native fetuin (F5)
glycan or linkage (i.e., glycoproteins forms βGNβGN and and in the glycans on the surface of the majority of
F2; Tn, A+ and sheep RBCs); thereby DGL-I may well bind animal RBCs tested is not incompatible with binding
the Forssman antigen, GalNAcα1-3GalNAc(α/β), known to by DGL-I (Table 1 and [4, 12]), and glycans with termi-
be present on sheep RBCs, but not on human or rabbit RBCs nal GlcNAc (GnGn) were bound by this lectin, but not
[62]. Meanwhile, DGL-II very weakly recognized the other two. On the other hand, GLL-II interacted with
asialoagalactofetuin (F2) and no interaction were observed with glycans containing exposed- (F4 form) or encrypted-IIβ
any other structure having terminal GalNAc. On the other hand, glycotope (H-type II determinant and rat RBC glycans)
there was an unresolved question as to the interaction of GLL-II and DGL-II could recognize encrypted IIβ glycotope (F5
with GalNAc: on one hand, asialosubmaxillary bovine mucin form, H-type II determinant and pig RBC glycans).
 Glycoconj J

Sialic acids the other hand, α1-2-linked fucose as present on H-type

histo-blood group determinants was bound by GLL-II,
Many lectins, whether from plants and animals, are known to with a higher affinity seen when a IIβ glycotope-bond
bind sialic acids [63–69]. In the present study we observed was present [6]. There seems to be some differentiation
DGL-I and DGL-II bound untreated fetuin (F5), but not native in the binding preferences of GLL-II and DGL-II towards
transferrin suggesting that α2-3-linked Neu5Ac (as found on α1,2-Fuc, i.e., the branching in glycans having Fucα1-2 as
fetuin but not on transferrin) could be a ligand of the in A and B histo-blood groups seems to prevent the recog-
D. grandiflora lectins. The same behavior was observed for nition by DGL-II but not for GLL-II (apart from the pres-
Triticum vulgaris lectin (WGA) with respect those two glyco- ence of GalNAc in A RBCs, which had decreased GLL-II
proteins [16], and in agreement with our results, previous data binding activity). Also, the GLL-II minimum agglutinating
did not indicate inhibition of DGL-I binding in presence of concentration towards O erythrocytes was 7 μg/mL, while
apotransferrin [10]. It was observed N-acetylneuraminic acid in the concentration range used in the present study we
inhibited the DGL-I activity on rabbit RBCs at concentration observed non-agglutinating DGL-II activity towards this
of 8.3 mM, which is, respectively, half or four times the in- type of RBCs and Melgarejo et al. observed a minimun
hibitory concentration of Glc and Man [10]. DGL-I aggluti- agglutinating DGL-II concentration of ~ 152 μg/mL [5].
nated horse and dog RBCs showing specific titers as high as Furthermore, aalthough both GLL-II and DGL-II showed
those obtained on rabbit RBCs [38]. The combined previous preference for the H-type II over other H-type groups, the
data suggest that horse RBCs contain large amount of Siaα2- GLL-II binding with H-type II-BSA neoglycoconjugate is
3Gal (primarily as Neu5Gc) and that dog RBCs express 10 times stronger than DGL-II. [5, 6]. On the other hand,
NeuAcα2-3Gal [58]; pig RBCs also have sialylated N- the presence of α1-2Fuc in the B histo-blood group re-
glycans (Supplemental Table S3), contain both NeuAc- and duces DGL-I activity on B RBCs as compared to rabbit
NeuGc-capped gangliosides [64] and react with an mAb [38] and porcine RBCs carrying the unsubstituted disac-
against α2-3-linked NeuAc [65], while expressing α-Gal charide Galα1-3Galβ1-R.
glycotopes at a level that is 250 fold lower than that on rabbit
red blood cells. Thus, we conclude that the DGL-II interaction Man3GlcNAc2 core (MM)
with native fetuin and pig RBCs could be due to the recogni-
tion of α2-3-linked Neu5Ac, while recognition of pig RBCs Only DGL-I recognized the MM transferrin glycoform,
by DGL-I could due to both αGal and α2-3Sia glycotopes. Manα1-6(Manα1-3)Manβ1-4GlcNAcβ1-4GlcNAcβ1-Asn-
On the other hand, GLL-II could agglutinate rat, but to X-Ser/Thr. This structure contains a Manα1-3(Manα1-6)Man
lesser extent than B and O, RBCs [6]. The agglutination of motif previously shown to be recognized by DGL-I
murine RBCs has been correlated with the presence of 9-O- (Ka = 122 M−1 × 10−4; complete inhibition with 12 μM of this
Ac-NeuAc [57, 64, 70] and it has observed some lectins ex- trisaccharide) [12] as well as by other Diocleae type-I lectins
hibited preference for binding this sialic acid type, e.g., the including those from Canavalia and Cratylia [8]. As discussed
lectin from crab blood [reviewed in 70]. As mentioned above, above, the present and previous studies [8, 9, 12, 13] show that
GLL-II could recognize external/internal IIβ glycotope; thus, DGL is sensitive to the nature of the linkage and/or saccharides
its interaction with the F5 glycoform and with rat RBCs could residues adjacent to the trimannoside moiety (i.e., elongation of
be due to the recognition of internal LacNAc, with terminal the mannose residues). On the other hand, DGL-II and GLL-II
sialic acids diminishing the binding. Besides, the presence of activities were not inhibited by Man [5, 6] and they anyway
terminal α2-6-linked Neu5Ac (as in native transferrin) show preference for O-glycans (Fig. 2).
completely prevented the recognition by all three Diocleae
lectins, similar to the case of galectin LGALS1 [71]. Gn2Man3 core

Fucose glycotopes Biantennary complex N-glycans can present the region

GlcNAcβ1-2(GlcNAcβ1-2)Manα1-3(Manα1-6)Man
Core α1-6-fucose is a widespread feature of the N-glycans (Gn2Man3) and the relative binding activities of Diocleae
in invertebrates and vertebrates [72]. However, it appears lectins for this core, in turn, have been shown to directly cor-
that none of the lectins studied here interact with this moi- relate with their histamine release activities from rat peritoneal
ety; indeed, α1-6-fucosylation may reduce the DGL-I in- mast cells [8]. Gn2Man3 is either capped or exposed in six of
teraction with N-glycans as suggested by the comparison the seven N-glycoforms used in the present study, and their
of the results with GnGnF6 and GnGn (Fig. 2). Also, the recognition by DGL-I depended on the surrounding residues
presence of α-linked Fuc to GlcNAc as in Lea or Ley de- (Fig. 2 and Supplemental Fig. S1); both the GnGn and
terminants could block the interaction with GLL-II, DGL- GnGnF6 glycoforms with terminal GlcNAcβ1-2Man being
II and CEL-II (Canavalia ensiformis lectin II) [5, 6]. On bound by this lectin, whereas galactosylated forms were not.
Glycoconj J

Hemagglutination inhibition and isothermal titration micro- revealed that proper contacts between DGL-I and the
calorimetry (ITC) experiments demonstrated substantial dif- Gn2Man3 region also are prevented due to key amino acid
ference in relative affinities of distinct Diocleae lectins for differences at residues 226 (Thr in ConA, Gly in DGL) and
Gn2Man3 and showing inhibitory concentrations in a range 168 (Ser in ConA, Asn in DGL), and a shift in the backbone of
of 6–790 mM, depending on the lectin [8, 9]. For example, it residues 222–227 [13].
was necessary a Gn2Man3 concentration of 400 mM for
inhibiting the rabbit RBC agglutination by DGL-I, while
6 mM for Canavalia ensiformis lectin type-I (ConA) [8]. Lectin specificity maps
DGL-I showed reduced affinities for the oligosaccharide
Gn2Man3 (Ka = 4.7 M−1 × 10−4) in comparison with ConA On the basis of our combined approach of using dot blot and
(Ka = 140 M−1 × 10−4) [9], despite their 81% amino acid se- lectin agglutinating activity, we show both the differences and
quence identity [13–15]. The dissimilarity observed in the overlaps in specificity between three Diocleae lectins. Taken
binding specificities of these two type-I lectins for Gn2Man3 together, the present and previous studies can be combined to
has been explained by the relatively unfavorable contact inter- suggest functional maps (Fig. 3) of the carbohydrate recogni-
actions of the β1,2GlcNAc residue on the α1,6 arm of the tion for DGL-I, DGL-II, and GLL-II. These maps illustrate the
complex carbohydrate in DGL-I [8, 9, 12, 13]. X-ray studies impact of modification of basic glycotopes enhancing,

Fig. 3 Lectin specificity maps. The left hand panels show the lectin of the most active ligands of DGL-I, trimannoside is exposed or
interaction with glycotopes Tα and IIβ. a Presence of exposed α1,3- encrypted in all N-glycoforms we used in the present study. Its interaction
linked Gal interferes with DGL-I binding to N-glycans but not to glyco- with the lectin is influenced by the surrounding sugar residues. Green
lipids (eg. αGal antigen on rabbit RBC glycolipids). Other glycotopes residues allow or enhance interaction; red residues decrease or not allow
which show or not interaction with the lectins are shown on the right hand interaction; word Bnone^: absence of more exposed residue(s)
panel: b Data from [38]. c Data from [8, 9]. d Data from [5, 6]. Below, one
 Glycoconj J

permitting, or inhibiting recognition, but obviously can be- 11. Ramos, M.V., Moreira, R.A., Cavada, B.S., de Oliveira, J.T.A.,
Rouge, P.: Interaction of lectins from the sub-tribe Diocleinae with
come even more precise in the future.
specific ligands. R. Bras. Fisiol Veg. 8, 193–199 (1996)
12. Gupta, D., Oscarson, S., Raju, T.S., Stanley, P., Toone, E.J., Brewer,
Acknowledgments We thank Dr. R.A. Moreira (University of Fortaleza, C.F.: A comparison of the fine saccharide-binding specificity of
Brazil) for his kind gift of D. grandiflora seeds; Dr.J.L. Fernández Dioclea Grandiflora lectin and Concanavalin a. Eur. J. Biochem.
(Instituto de Ciencias Naturales, Universidad Nacional de Colombia) 242, 320–326 (1996)
for species identification and Dr. Wilson’s staff at the Universität für 13. Rozwarski, D.A., Swami, B.M., Brewer, C.F., Sacchettini, J.C.:
Bodenkultur (BOKU, Vienna, Austria) for training and collaboration. Crystal structure of the lectin from Dioclea Grandiflora complexed
Financial support was provided by the Chemistry Department at the with core trimannoside of asparagine-linked carbohydrates. J. Biol.
Universidad Nacional - Bogotá, Colombia and COLCIENCIAS with Chem. 273, 32818–32825 (1998)
project Lamiaceae Lectin Structure (Grant 110148925106). 14. Richardson, M., Campos, F.D., Moreira, R.A., Ainouz, I.L.,
COLCIENCIAS doctoral student grant (Grant 617 – 0656) facilitated Begbie, R., Watt, W.B., Pusztai, A.: The complete amino acid se-
the internship of T.M.Cortázar at the BOKU. quence of the major alpha subunit of the lectin from the seeds of
Dioclea Grandiflora (Mart). Eur. J. Biochem. 144, 101–111 (1984)
Compliance with ethical standards 15. Ainouz, I.L., Moreira, R.A., Campos, F.D.A.P., Richardson, M.,
Begbie, R., Stewart, J.C., Watt, W.B., Pusztai, A.: The isolation
Conflicts of interest The authors declare that they have no conflict of and amino acid sequence of the β- and γ-subunits of the lectin from
interest. the seeds of Dioclea Grandiflora. Phytochemistry. 26, 1435–1440
(1987)
16. Iskratsch, T., Braun, A., Paschinger, K., Wilson, I.B.H.: Specificity
Ethical approval This article does not contain any studies with human
analysis of lectins and antibodies using remodeled glycoproteins.
participants or animals performed by any of the authors.
Anal. Biochem. 386, 133–146 (2009)
17. Hermanson, G.T., Mallia, A.K., Smith, P.: Activation Methods. In:
Immobilized Affinity Ligand Techniques, pp. 88–91. Academic
References Press, San Diego (1992)
18. Smith, P.K., Krohn, R.I., Hermanson, G.T., Mallia, A.K., Gartner,
F.H., Provenzano, M.D., Fujimoto, E.K., Goeke, N.M., Olson, B.J.,
1. Dan, X., Liu, W., Ng, T.B.: Development and applications of lectins Klenk, D.C.: Measurement of protein using bicinchoninic acid.
as biological tools in biomedical research. Med. Res. Rev. 36, 221– Anal. Biochem. 150, 76–85 (1985)
247 (2016) 19. Schägger, H., von Jagow, G.: Tricine-sodium dodecyl sulfate-
2. André, S., Kaltner, H., Manning, J.C., Murphy, P.V., Gabius, H.J.: polyacrylamide gel electrophoresis for the separation of proteins
Lectins: getting familiar with translators of the sugar code. in the range from 1 to 100 kDa. Anal. Biochem. 166, 368–379
Molecules. 20, 1788–1823 (2015) (1987)
3. de Queiroz, L.P., Pastore, J.F., Cardoso, D., Snak, C., de C Lima, 20. Laemmli, U.K.: Cleavage of structural proteins during the assembly
A.L., Gagnon, E., Vatanparast, M., Holland, A.E., Egan, A.N.: A of the head of bacteriophage T4. Nature. 227, 680–685 (1970)
multilocus phylogenetic analysis reveals the monophyly of a 21. Wu, A.M., Duk, M., Lin, M., Broadberry, R.E., Lisowska, E.:
recircumscribed papilionoid legume tribe Diocleae with well- Identification of variant glycophorins of human red cells by
supported generic relationships. Mol. Phylogenet. Evol. 90, 1–19 lectinoblotting: application to the mi.III variant that is relatively
(2015) frequent in the Taiwanese population. Transfusion. 35, 571–576
4. Moreira, R.A., Barros, A.C.H., Stewart, J.C., Pusztai, A.: Isolation (1995)
and characterization of a lectin from the seeds of Dioclea 22. Iglesias, J.L., Lis, H., Sharon, N.: Purification and properties of a D-
Grandiflora (Mart.) Planta (Heidelh.) 158, 63–69 (1983) galactose/N-acetyl-D-galactosamine-specific lectin from Erythrina
5. Melgarejo, L.M., Vega, N., Pérez, G.: Isolation and characterization Cristagalli. Eur. J. Biochem. 123, 247–252 (1982)
of novel lectins from Canavalia Ensiformis DC and Dioclea 23. Hirohashi, S., Clausen, H., Yamada, T., Shimosato, Y., Hakomori,
Grandiflora Mart. Ex Benth. seeds. Braz. J. Plant Physiol. 17, S.I.: Blood group a cross-reacting glycotope defined by monoclonal
315–324 (2005) antibodies NCC Lu-35 and 81 expressed in cancer of blood group O
6. Almanza, M., Vega, N., Pérez, G.: Isolating and characterising a or B individuals. Its identification as Tn antigen. Proc. Natl. Acad.
lectin from Galactia Lindenii seeds that recognises blood group H Sci. 82, 7039–7043 (1985)
determinants. Arch. Biochem. Biophys. 429, 180–190 (2004) 24. Vega, N., Pérez, G.: Isolation and characterisation of a Salvia
7. Burkart, A.E.: El género Galactia (Leguminosae—Phaseoleae) en B o g o t e n si s s e e d l e c t i n s p e c i f i c fo r t h e Tn a n t i g e n .
Sudamérica con especial referencia a la Argentina y países vecinos. Phytochemistry. 67, 347–355 (2006)
Darwin. 16, 663–796 (1971) 25. Sumner, J.B., Howell, S.F.: The identification of the hemagglutinin
8. Dam, T.K., Cavada, B.S., Grangeiro, T.B., Santos, C.F., de Sousa, of the jack bean with Concanavalin a. J. Bacteriol. 32, 227–237
F.A.M., Oscarson, S., Brewer, C.F.: Diocleinae lectins are a Group (1936)
of Proteins with conserved binding sites for the Core Trimannoside 26. Moreira, R.A., Cavada, B.S.: Lectin from Canavalia Brasiliensis
of asparagine-linked oligosaccharides and differential specificities (Mart.). Isolation, characterization and behavior during germina-
for complex carbohydrates. J. Biol. Chem. 273, 12082–12088 tion. Biol. Plantarum. 26, 113–120 (1984)
(1998) 27. Oliveira, J.T.A., Cavada, B.S., Moreira, R.A.: Isolation and partial
9. Dam, T.K., Roy, R., Das, S.K., Oscarson, S., Brewer, C.F.: Binding characterization of a lectin from Cratylia floribunda Mart. Seeds.
of multivalent carbohydrates to Concanavalin a and Dioclea Rev. Bras. Botânica. 14, 63–68 (1991)
Grandiflora lectin. J. Biol. Chem. 275, 14223–14230 (2000) 28. Vasconcelos, I.M., Cavada, B.S., Moreira, R.A., Oliveira, J.T.A.:
10. Ramos, M.V., Moreira, R.A., Oliveira, J.T., Cavada, B.S., Rouge, Purification and partial characterization of a lectin from the seeds of
P.: The carbohydrate binding specificity and molecular modelling Dioclea Guianensis. J. Food Biochem. 15, 137–154 (1991)
of Canavalia Maritima and Dioclea Grandiflora lectins. Mem. Inst. 29. Pérez, G., Perez, C., Cavada, B.S., Moreira, R.A., Richardson, M.:
Oswaldo Cruz. 91, 761–766 (1996) Comparison of the amino acid sequences of the lectins from seeds
Glycoconj J

of Dioclea Lehmanni and Canavalia Maritima. Phytochemistry. 30, 45. Green, E.D., Adelt, G., Baenziger, J.U., Wilson, S., Van Halbeek,
1619–1621 (1991) H.: The asparagines linked oligosaccharides on bovine fetuin: struc-
30. Moreira, R.A., Cordeiro, E.F., Ramos, M.V., Grangeiro, T.B., tural analysis of N –glycanase released oligosaccharides by 500-
Martins, J.L., Oliveira, J.T.A., Cavada, B.S.: Isolation and partial megahertz 1H NMR spectroscopy. J. Biol. Chem. 263, 18253–
characterization of a lectin from seeds of Dioclea Violacea. Braz. J. 18268 (1988)
Plant Physiol. 8, 23–29 (1996) 46. Ravn, V., Dabelsteen, E.: Tissue distribution of histo-blood group
31. Moreira, R.A., Monteiro, A.C.O., Horta, A.C.G., Oliveira, J.T.A., antigens. APMIS. 108, 1–28 (2000)
Cavada, B.S.: Isolation and characterization of Dioclea Altissima 47. Evans, S.V., MacKenzie, C.R.: Characterization of protein–glyco-
Var. Megacarpa seed lectin. Phytochemistry. 46, 139–144 (1997) lipid recognition at the membrane bilayer. J. Mol. Recognit. 12,
32. Correia, J.L., do Nascimento, A.S., Cajazeiras, J.B., Gondim, A.C., 155–168 (1999)
Pereira, R.I., de Sousa, B.L., Coelho de Silva, A.L., Garcia, W., 48. Himbert, S., Alsop, R.J., Rose, M., Hertz, L., Dhaliwal, A., Moran-
Teixeira, E.H., do Nascimento, K.S., da Rocha, B.A., Nagano, Mirabal, J.M., Verschoor, C.P., Bowdish, D.M., Kaestner, L.,
C.S., Sampaio, A.H., Cavada, B.S.: Molecular characterization Wagner, C., Rheinstädter, M.C.: The molecular structure of human
and tandem mass spectrometry of the lectin extracted from the seeds red blood cell membranes from highly oriented. Solid Supported
of Dioclea Sclerocarpa. Molecules. 16, 9077–9089 (2011) Multi-Lamellar Membranes. Sci. Rep. 7, 39661 (2017). https://doi.
33. do Nascimento, A.S., Gondim, A.C., Cajazeiras, J.B., Correia, J.L., org/10.1038/srep39661
Pires Ade, F., do Nascimento, K.S., da Silva, A.L., Nagano, C.S., 49. Anstee, D.J.: Blood group-active surface molecules of the human
Assreuy, A.M., Cavada, B.S.: Purification and partial characteriza- red blood cell. Vox Sang. 58, 1–20 (1990)
tion of a novel lectin from Dioclea Lasiocarpa Mart seeds with 50. Cao, Y., Merling, A., Karsten, U., Goletz, S., Punzel, M., Kraft, R.,
vasodilator effects. J. Mol. Recognit. 25, 657–664 (2012) Butschak, G., Schwartz-Albiez, R.: Expression of CD175 (Tn),
34. Pinto-Júnior, V.R., de Santiago, M.Q., Osterne, V.J., Correia, J.L., CD175s (sialosyl-Tn) and CD176 (Thomsen-Friedenreich
Pereira-Júnior, F.N., Cajazeiras, J.B., de Vasconcelos, M.A., antigen) on malignant human hematopoietic cells. Int. J. Cancer.
Teixeira, E.H., do Nascimento, A.S., Miguel, T.B., Miguel Ede, 123, 89–99 (2008)
C., Sampaio, A.H., do Nascimento, K.S., Nagano, C.S., Cavada, 51. Podbielska, M., Fredriksson, S.A., Nilsson, B., Lisowska, E.,
B.S.: Purification, partial characterization and immobilization of a Krotkiewski, H.: ABH blood group antigens in O-glycans of hu-
mannose-specific lectin from seeds of Dioclea Lasiophylla. man glycophorin a. Arch. Biochem. Biophys. 429, 145–153 (2004)
Molecules. 18, 10857–10869 (2013) 52. Koizumi, T., Matsumoto-Takasaki, A., Nakada, H., Nakata, M.,
35. Pinto-Junior, V.R., Correia, J.L., Pereira, R.I., Pereira-Junior, F.N., Fujita-Yamaguchi, Y.: Preparation of asialo-agalacto-glycophorin
Santiago, M.Q., Osterne, V.J., Madeira, J.C., Cajazeiras, J.B., a for screening of anti-Tn antibodies. Bio Sci Trends. 4, 308–311
Nagano, C.S., Delatorre, P., Assreuy, A.M., do Nascimento, K.S., (2010)
Cavada, B.S.: Purification and molecular characterization of a novel
53. Ogawa, H., Galili, U.: Profiling terminal N-acetyllactosamines of
mannose-specific lectin from Dioclea Reflexa seeds with inflam-
glycans on mammalian cells by an immuno-enzymatic assay.
matory activity. J. Mol. Recognit. 29, 134–142 (2016)
Glycoconj. J. 23, 663–674 (2006)
36. Cavada, B.S., Grangeiro, T.B., Ramos, M.V., Cordeiro, E.F.,
54. Doucet, J., Gao, Z.H., MacLaren, L.A., McAlister, V.C.:
Oliveira, J.T.A., Moreira, R.A.: Isolation and partial characteriza-
Modification of xenoantigens on porcine erythrocytes for
tion of a lectin from Dioclea Rostrata Benth seeds. R. Bras. Fisiol
xenotransfusion. Surgery. 135, 178–186 (2004)
Veg. 8, 31–36 (1996)
37. Pérez, G., Hernandez, M., Mora, E.: Isolation and characterization 55. Tan, Y., Gong, F., Li, S., Ji, S., Lu, Y., Gao, H., Xu, H., Zhang, Y.: A
of a lectin from the seeds of Dioclea Lehmanni. Phytochemistry. 29, new profile of terminal N-acetyllactosamines glycans on pig red
1745–1719 (1990) blood cells and different expression of alpha-galactose on sika deer
38. Pérez, G.: Isolation and characterization of a novel lectin from red blood cells and nucleated cells. Glycoconj. J. 27, 427–433
Dioclea Lehmanni (Fabaceae) seeds. Int. J. Biochem. Cell Biol. (2010)
30, 843–853 (1998) 56. Momoi, M., Yamakawa, T.: Glucosamine containing
39. Sierra, A.Y., Pérez, G.: Extracción, purificación y caracterización de Sphingoglycolipids from sheep erythrocytes. J. Biochem. 84,
dos lectinas en semillas de Dioclea sericea. Rev. Acad. Colomb. 317–325 (1978)
Cienc. 23, 445–454 (1999) 57. Ito, T., Suzuki, Y., Mitnaul, L., Vines, A., Kida, H., Kawaoka, Y.:
40. Cavada, B.S., Ramos, M.V., Cordeiro, E.F., Grangeiro, T.B., Receptor specificity of influenza a viruses correlates with the ag-
Oliveira, J.T.A.: Urano de Carvalho. A.F.F., Moreira, R.A.: purifi- glutination of erythrocytes from different animal species. Virol.
cation and partial characterization of a lectin from Dioclea Virgata 227, 493–499 (1997)
Benth seeds. R. Bras. Fisiol Veg. 8, 37–42 (1996) 58. Dam, T.K., Gerken, T.A., Cavada, B.S., Nascimento, K.S., Moura,
41. Rangel, T.B., Assreuy, A.M., Pires, A.F., Carvalho, A.U., T.R., Brewer, C.F.: Binding studies of α-GalNAc-specific lectins to
Benevides, R.G., Simões, R.C., Silva, H.C., Bezerra, M.J., the α-GalNAc (Tn-antigen) form of porcine submaxillary mucin
Nascimento, A.S., do Nascimento, K.S., Nagano, C.S., Sampaio, and its smaller fragments. J. Biol. Chem. 282, 28256–28263 (2007)
A.H., Delatorre, P., Rocha, B.A., Fernandes, P.M., Cavada, B.S.: 59. Wu, A.M., Liu, J.H., Gong, Y.P., Li, C.C., Chang, E.T.: Multiple
Crystallization and characterization of an inflammatory lectin pu- recognition systems adopting four different glycotopes at the same
rified from the seeds of Dioclea Wilsonii. Molecules. 16, 5087– domain for the Agaricus bisporus agglutinin-glycan interactions.
5103 (2011) FEBS Lett. 584, 3561–3566 (2010)
42. Carrington, D.M., Auffret, A., Hanke, D.E.: Polypeptide ligation 60. Piller, V., Piller, F., Cartron, J.P.: Comparison of the carbohydrate-
occurs during post- translational modification of concanavalin a. binding specificities of seven N-acetyl-D-galactosamine-
Nature. 313, 64–67 (1985) recognizing lectins. Eur. J. Biochem. 191, 461–466 (1990)
43. Satomi, Y., Shimonishi, Y., Takao, T.: N-glycosylation at Asn491 in 61. Leppänen, A., Stowell, S., Blixt, O., Cummings, R.D.: Dimeric
the Asn-Xaa-Cys motif of human transferrin. FEBS Lett. 576, 51– galectin-1 binds with high affinity to alpha2,3-sialylated and non-
56 (2004) sialylated terminal N-acetyllactosamine units on surface-bound ex-
44. Spiro, R.G., Bhoyroo, V.D.: Structure of the O-glycosidically tended glycans. J. Biol. Chem. 280, 5549–5562 (2005)
linked carbohydrateunits of fetuin. J. Biol. Chem. 249, 5704– 62. Yamamoto, M., Cid, E., Yamamoto, F.: Crosstalk between ABO
5717 (1974) and Forssman (FORS) blood group systems: FORS1 antigen
 Glycoconj J

synthesis by ABO gene-encoded glycosyltransferases. Sci. Rep. 7, parameters of lectins by using surface plasmon resonance. Anal.
41632 (2017). https://doi.org/10.1038/srep41632 Biochem. 274, 203–210 (1999)
63. Ahmad, N., Gabius, H.J., Kaltner, H., André, S., Kuwabara, I., Liu, 68. Varki, A., Gagneux, P.: Multifarious roles of sialic acids in immu-
F.T., Oscarson, S., Norberg, T., Brewer, C.F.: Thermodynamic bind- nity. Ann. N. Y. Acad. Sci. 1253, 16–36 (2012)
ing studies of cell surface carbohydrate glycotopes to galectins-1, 69. Zou, Z., Chastain, A., Moir, S., Ford, J., Trandem, K., Martinelli, E.,
−3, and −7: evidence for differential binding specificities. Can. J. Cicala, C., Crocker, P., Arthos, J., Sun, P.D.: Siglecs facilitate HIV-1
Chem. 80, 1096–1104 (2002) infection of macrophages through adhesion with viral sialic acids.
64. Suzuki, Y., Suzuki, N., Michi, H., Matsumoto, M.: Characterization PLoS One. 6(9), e24559 (2011). https://doi.org/10.1371/journal.
of gangliosides of porcine erythrocyte membranes: occurrence of pone.0024559
ganglioside GD3 as major ganglioside. Lipids. 20, 588–593 (1985) 70. Rogers, G.N., Herrler, G., Paulson, J.C., Klenk, H.D.: Influenza C
65. Thorpe, S., Feizi, T.: Species differences in the expression of car- virus uses 9-0-acetyl-N-acetylneuraminic acid as a high affinity
bohydrate differentiation antigens on mammalian blood cells re- receptor determinant for attachment to cells. J. Biol. Chem. 261,
vealed by immunofluorescence with monoclonal antibodies. 5947–5951 (1986)
Biosci. Rep. 4, 673–685 (1984) 71. Rapoport, E.M., Kurmyshkina, O.V., Bovin, N.V.: Mammalian
66. Schauer, R., Kamerling, J.P.: Chemistry, Biochemistry and Biology galectins: structure, carbohydrate specificity, and functions.
of Sialic Acids. In: New Comprehensive Biochemistry, Biochem. Mosc. 73, 393–405 (2008)
Glycoproteins II, pp. 243–372. Elsevier, New York (1997)
72. Wilson, I.B.H.: Glycosylation of proteins in plants and inverte-
67. Haseley, S.R., Talaga, P., Kamerling, J.P., Vliegenthart, J.F.G.:
brates. Curr. Opin. Struct. Biol. 12, 569–577 (2002)
Characterization of the carbohydrate binding specificity and kinetic