Glycoconjugate Journal

, 25:741

Antiproliferative effect of T/Tn specific Artocarpus lakoocha agglutinin (ALA) on human leukemic cells (Jurkat, U937, K562) and their imaging by QD-ALA nanoconjugate

Authors

  • Urmimala Chatterjee
    • Department of Biological ChemistryIndian Association for the Cultivation of Science
    • Department of Anatomy, Physiology and BiochemistrySwedish University of Agricultural Science
  • Partha Pratim Bose
    • Department of Biological ChemistryIndian Association for the Cultivation of Science
    • Department of Biochemistry and Organic ChemistryUppsala University
  • Sharmistha Dey
    • Department of BiophysicsAll India Institute of Medical Sciences
  • Tej P. Singh
    • Department of BiophysicsAll India Institute of Medical Sciences
    • Department of Biological ChemistryIndian Association for the Cultivation of Science
    • Department of BiotechnologyWest Bengal University of Technology
Article

DOI: 10.1007/s10719-008-9134-8

Cite this article as:
Chatterjee, U., Bose, P.P., Dey, S. et al. Glycoconj J (2008) 25: 741. doi:10.1007/s10719-008-9134-8

Abstract

T/Tn specificity of Artocarpus lakoocha agglutinin (ALA), isolated from the seeds of A. lakoocha (Moraceae) fruit and a heterodimer (16 kD and 12 kD) of molecular mass 28 kD, was further confirmed by SPR analysis using T/Tn glycan containing mammalian glycoproteins. N-terminal amino acid sequence analysis of ALA showed homology at 15, 19–21, 24–27, and 29 residues with other lectin members of Moraceae family viz., Artocarpus integrifolia (jacalin) lectin, Artocarpus hirsuta lectin, and Maclura pomifera agglutinin. It is mitogenic to human PBMC and the maximum proliferation was observed at 1 ng/ml. It showed an antiproliferative effect on leukemic cells, with the highest effect toward Jurkat cells (IC50 13.15 ng/ml). Synthesized CdS quantum dot-ALA nanoconjugate was employed to detect the expression of T/Tn glycans on Jurkat, U937, and K562 leukemic cells surfaces as well as normal lymphocytes by fluorescence microscopy. No green fluorescence was observed with normal lymphocytes indicating that T/Tn determinants, which are recognized as human tumor associated structures were cryptic on normal lymphocyte surfaces, whereas intense green fluorescent dots appeared during imaging of leukemic cells, where such determinants were present in unmasked form. The above results indicated that QD-ALA nanoconjugate is an efficient fluorescent marker for identification of leukemic cell lines that gives rise to high quality images.

Keywords

ALAArtocarpus lakoocha agglutininSurface plasmon resonanceLeukemic cellsQuantum dotNanoconjugate

Abbreviations

ALA

Artocarpus lakoocha agglutinin

BSM

Bovine submandibular gland mucin

CdS

Cadmium sulphide

ELLSA

Enzyme - linked lectinsorbent assay

ESI-MS

Electron spray ionization mass spectrometry

FPLC

Fast protein liquid chromatography

HBS

Hepes buffered saline

HEPES

N-(2-hydoxyethyl) piperizine-N′-(2-hydroxypropane sulfonic acid)

HSM

Hamster submaxillary mucin

PBMC

Peripheral blood mononuclear cells

PVDF

Polyvinylidene difluoride

Q-ToF

Quadrupole-time of flight

QD

Quantum dot

SDS-PAGE

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis

2 ME

2-Mercaptoethanol

1 Introduction

Lectins comprise a structurally diverse class of proteins or glycoproteins other than antibodies or enzymes that bind specifically and reversibly to mono- and oligosaccharides [1]. They are ubiquitous in the biosphere and have been isolated from viruses, bacteria, fungi, plants and animals. Because of their well defined specificity they serve as valuable reagents in studies of biochemistry, cell biology, hematology, immunology, glycobiology, and oncology and have enormous applications in biomedical researches including cancer research [2]. Lectins aid to differentiate between the malignant and normal cells based on their agglutination pattern. This is due to altered glycosylation on cell surface associated with malignancy, its progression and metastasis [3, 4]. They can be employed for the detection of glycan changes in certain disease processes which involve fucosylation, increased sialylation and increased branching of complex carbohydrates. Recognition of these altered structural profiles of glycans by lectins provides valuable disease biomarkers [58]. Recent development in quantum dot (QD) nanotechnology have resulted in the introduction of new fluorescent immunocytochemical probes, in which QDs are covalently coupled to lectins. These were used for ultrasensitive biological imaging and analysis using fluorescent microscopy and/or flow cytometry [911].

Previously, we isolated a lectin from the seeds of Artocarpus lakoocha fruit, a plant belonging to Moraceae family, and its carbohydrate specificity was defined against monosaccharides and several T related disaccharides [1214]. Further examination of its glycan affinity at macromolecular level by enzyme-linked lectinsorbant assay and binding-inhibition assay using mammalian glycotopes in macromolecules revealed its binding specificity to tumor-associated carbohydrate antigens GalNAcα1→Ser/Thr (Tn) and Galβ1-3GalNAcα1→Ser/Thr (Tα). It hardly cross-reacted with common glycotopes on glycoproteins, including ABH blood group antigens, Galβ1-3/4GlcNAc determinants, T/Tn covered by sialic acids and N-linked serum glycoproteins [15]. However, no biological property of this lectin was reported till now. The present study reports purification of this lectin by affinity repulsion chromatographic technique, some of its biological properties and the binding specificity toward T/Tn containing glycoproteins using surface plasmon resonance analysis. The present work also demonstrates a unique and simple technique based on cadmium sulphide nanoparticle (QD) tagged ALA, which was used as a sensitive probe for distinguishing leukemic cell lines, viz., Jurkat, U937 and K562 from normal lymphocytes.

2 Materials and methods

RPMI-1640 culture medium and fetal bovine serum (FBS) were purchased from GIBCO, USA. Hepes, sodium bicarbonate, Ficoll-Histopaque, 2-mercaptoethanol, penicillin, streptomycin, fungizone, glutamine, biotinamidocaproate-N-hydroxy-succinimide ester, antibiotin-HRP and trypan blue were purchased from Sigma, USA. [3H] Thymidine was the product of New England Nuclear, USA. Tissue culture plates (96 well), petri dishes, culture bottles were procured from Axygen, Sweden. Asialo BSM, Tn containing glycopeptides (MW < 3,000 Da) and T disaccharide (Galβ1→3GalNAc) were the kind gift of Prof. A. M. Wu, Chang-Gung University, Kwei-San, Taipe, Taiwan. Methyl-α-galactose (Me-α-Gal), Methyl-α-N-acetylgalactosamine (Me-α-GalNAc) were kindly obtained from Prof. N. Roy, Department of Biological Chemistry, Indian Association for the Cultivation of Science, Kolkata, India. Asialo glycophorin which has fifteen O-linked glycan of Tα (Galβ1→3GalNAc α1→Ser/Thr) lectin determinants and one N-linked carbohydrate residue was generously supplied by Prof. E. Lisowska (Institute of Immunology and Experimental Therapy, Wroclaw, Poland, and prepared according to Wu and Pigman [16]. Tn-glycophorin was prepared by removing galactose residue from asialo glycophorin by periodate oxidation and mild acid hydrolysis [17]. Fetuin and BSM (Sigma) were desialylated by 0.01M HCl at 800 C for 90 min. Small fragments and HCl were removed by extensive dialysis against H2O.

2.1 Purification of Artocarpus lakoocha agglutinin

ALA was purified by a combination of two methods [12, 13] as reported earlier [15]. In brief, the saline extract of the seeds (20% w/v) was precipitated by 40% saturated ammonium sulfate with vigorous stirring for 3 h at 4°C. The resulting precipitate was dissolved in a minimum volume of saline and was subjected to 0.2% (w/v) rivanol (6, 9-diamino-2-ethoxy acridine lactate) precipitation. After removal of excess rivanol by precipitation with KBr, the supernatant was extensively dialyzed against saline overnight at 4°C. The dialyzed solution after concentration by YM 10 membrane was subjected to affinity chromatography on melibiose-agarose column (20 cm × 1 cm). The unbound proteins were eluted with 10 mM PBS, pH 7.2 till A280 of effluents was less than 0.002. The bound protein was desorbed with deionized water following the principle of affinity repulsion chromatography [18]. The active fractions were mixed and after concentration by YM 10 membrane was stored −20°C with 0.02% NaN3 for further work. The minimum concentration of ALA for erythroagglutination was 3.57 ng/ml. Protein content of all the fractions were determined by Bradford method [19] using BSA as the standard.

2.2 Hemagglutination

The hemagglutinating activity of the purified agglutinin was determined by incubating a twofold serially diluted agglutinin solution (25 μl) in saline with an equal volume of 2% (v/v) heparinized normal human B erythrocytes suspension in saline for 30 min at room temperature. The normal erythrocytes were prepared according to Bhowal et al. [20]. Hemagglutination titer was defined as the reciprocal of the highest dilution showing visible hemagglutination.

2.3 PAGE, SDS-PAGE and molecular mass

Polyacrylamide gel electrophoresis (PAGE) under non-denaturing condition was performed on 0.75 mm slab gel (10%) in acidic buffer system (β alanine/acetic acid, pH 4.3) according to Reisfled et al. [21]. The protein bands were visualized by Coomassie brilliant blue G-250 staining followed by destaining in 5% acetic acid.

SDS-PAGE was done on 12% polyacrylamide gel in Tris/glycine buffer pH 8.3 according to the method of Laemmli [22]. The sample was heated with 1% SDS in the presence or absence of 2-ME for 5 min at 1000 C. The gel was stained with Coomassie brilliant blue, G-250. Molecular mass of the lectin was calculated according to the relative mobility with the Precision Plus Protein™ Standards.

The molecular mass of ALA was determined by gel filtration using Superose 6, 10/300 GL column in FPLC system (Amersham Biosciences, Sweden) at 4°C in TBS at a flow rate 24 ml/h. The molecular mass was calculated from the relative elution volume of different standard proteins of gel filtration kit (Sigma, USA). The molecular mass of ALA was also determined by ESI-MS using 5 μl/min flow rate of 100 pmol lectin solution (50:50:1; acetonitrile: water: formic acid) on Micromass Q-ToF Micro (Waters) mass spectrometer.

2.4 N-terminal amino acid sequencing

ALA was run on 12% SDS-PAGE and electroblotted onto polyvinylidene difluoride (PVDF) membrane at 100 mA constant current for 18 h using 10 mM 3-cyclohexylamino-1-propane-sulfonic acid (CAPS), pH 11.0, containing 10% MeOH as the blotting buffer [23]. After transfer of the protein band to PVDF, the membrane was rinsed several times with Milli Q water and then saturated with 100% MeOH for 2 s. The blot was next stained with 0.1% Amido black in 40% MeOH containing 1% AcOH for 1 min and destained with 50% MeOH. The membrane was washed with several changes of water, air dried and the protein band was excised.

The N-terminal amino acid sequence of the electroblotted ALA was done in a gas-phase protein sequencer (Shimadzu model PPSQ-21A) consisting of Edman reaction unit followed by HPLC and UV detection [24].

2.5 Glycan detection of ALA

ALA was electrophoresed on 10% SDS-PAGE gel and blotted onto nitrocellulose membrane, using transfer buffer 0.192 M glycine, 25 mM Tris pH 8.9, 20% (v/v) methanol at 100 mA constant current for 4 h [23]. The complete transfer of ALA bands was checked by staining with Ponceau S (0.2% Ponceau S in 1% acetic acid). The type of glycoform in ALA subunits was performed using DIG Glycan Differentiation Kit (Roche Applied Science) after transferring the lectin bands to NCM.

2.6 Binding studies of ALA with glycoproteins and glycans by surface plasmon resonance (SPR) analysis

The binding studies were carried out using the BIAcore 2000 SPR apparatus, (BIAcore AB, Uppsala, Sweden) at 25°C. The BIAcore apparatus is a biosensor-based system for real time specific interaction analysis [25]. The sensor chip CM5, surfactant P20, amine coupling kit containing N-hydroxysuccinimide (NHS), N-ethyl-N″-3(diethylaminopropyl) carbodimide (EDC) and ethanolamine hydrochloride used were supplied by BIAcore AB. After equilibration with 5 mM HEPES-buffered saline (pH 7.4) and 0.005% surfactant P20, the surface of the sensor chip was activated with a 1:1 mixture (70 μl) of 0.1 M NHS and 0.1 M EDC. ALA (200 μg/ml) was immobilized in 10 mM sodium acetate buffer (pH 4.8) at a flow rate of 5 μl/min for 30 min and unreacted groups were blocked by injection of 1.0 M ethanolamine (pH 8.5). The association rate constants were determined by passing the glycan solutions (5–20 nM) over the chip at a flow rate of 5 μl/min for 8 min. The dissociation rate constants were determined by passing HBS at a flow rate of 5 μl/min for 8 min and dissociation was followed by passing HBS at a flow rate of 5 μl/min for 10 min. After every cycle the sensor chip was regenerated by passing 100 mM HCl for 1 min. Kinetic parameters were calculated by BIA evaluation software version 3.0.

2.7 Isolation of human peripheral blood mononuclear cells

Blood from normal healthy individual taken in heparin was layered on Ficoll-Histopaque solution and centrifuged at 1,800 rpm for 30 min at room temperature. The PBMC were collected from the interface, washed thrice with RPMI-1640, and contaminating erythrocytes were depleted by hypotonic shock. The cells were suspended in RPMI-1640 supplemented with fetal bovine serum (10%), Hepes (25 mM), glutamine (200 μM), penicillin (100 IU/ml), streptomycin (100 μg/ml) and 0.5% fungizone [26]. The cellular viability was assessed by the trypan blue exclusion test. The percentage of cells that excluded the dye was enumerated in a hemocytometer under phase contrast microscope (Leitz, Diaplan).

2.8 Assay of mitogenic activity of ALA

The PBMC suspension (106 cells/ml, 0.1 ml) was cultured for 72 h at 37°C in humidified atmosphere of 5% CO2 (Heraus) with different concentrations of ALA (0.01, 0.1, 1, 10, 100 and 1,000 ng/ml) in 96 well tissue culture plates in triplicate [27]. The wells that received only PBMC suspension were used as negative control and those with PBMC suspension along with ConA (3 μg/ml) were treated as positive control. To each well of the plates incubated for 72 h radioactive precursor, 1 μCi [H]3 thymidine was added and the plates were incubated for 18 h. The cells were harvested onto GFC membrane by a cell harvester. The incorporated radioactivity was measured by liquid scintillation counter [28].

2.9 Antiproliferative activity of ALA on leukemic cell lines

The human leukemic cell lines Jurkat, U937 and K562 were obtained from National Centre for Cell Science, Pune. All cell lines were cultured and maintained in RPMI 1640 as described before. The cells attained 70% confluency within 48 h. The antiproliferative activity of ALA in vitro was determined as follows [29]. Cells (1 × 104/ ml) in their exponential growth phase were seeded into each well of 96 well culture plates (NUNC) and incubated for 3 h. Thereafter ALA (0.4, 4 and 40 ng/ml) was added and incubation was carried out for another 48 h. Radioactive precursor, 1 μCi [H]3 thymidine was added to each well and incubated for 18 h. The cells were harvested onto GFC membrane by a cell harvester. The incorporated radioactivity was measured by liquid scintillation counter.

2.10 Synthesis of ALA conjugated cadmium sulphide nanoparticle

CdS-nanoparticle tagged ALA was prepared by arrested precipitation of CdS nanocrystals at 18°C in solution phase using ALA as the colloidal stabilizer. CdS nanocrystals were prepared as follows. 0.9 mg of CdCl2 was dissolved in 100 ml (0.05 mM) of deionized water with vigorous stirring. 1 ml of 0.1 mM ALA solution (in deionized water) was added to the 1 ml stirred solution of CdCl2 prepared above and the mixture was allowed to stir gently for another 15 min. Thereafter 20 μl of Na2S, 9H2O (0.1 M) solution in tris-buffer (20 mM, pH 8.2) was added to the previous solution in a stirring condition. The solution turned faint green due to in situ formation of lectin stabilized CdS quantum dot (QD). The formation of lectin bound CdS nanoparticle was ensured by the photoluminescence spectra, Transmission electron microscopy (TEM), Energy Dispersive X-ray analysis (EDX) and fluorescence microscopy. The activity of QD-ALA was determined by hemagglutination with human erythrocytes along with ALA.

2.11 Transmission electron microscopy (TEM)

Transmission electron microscopy (TEM) was carried out using a JEOL JEM-2010 electron microscope to observe finer morphological details of the ALA - tagged quantum dots. The experiment was performed using a small amount of the aforementioned solution in tris-buffer on carbon-coated copper grid (300 mesh) by slow evaporation and subsequent vacuum drying at 18°C for 2 days. Images were taken at an accelerating voltage of 200 kV.

2.12 Fluorescence microscopy of the interaction of QD-ALA conjugate with cells

To examine the attachment of QD-ALA to various cells such as Jurkat, U937, K562 and normal PBMC, the cells were washed twice with 10 mM phosphate buffer saline (PBS), pH 7.4 and fixed with 4% paraformaldehyde in PBS for 10 min. The fixed cells were then blocked by 1% (w/v) BSA in PBS for 20 min and were separately incubated with 10 μl of QD-ALA solution for 1 h. Following incubation the cells were washed thoroughly twice with PBS. A drop of QD-ALA attached cell suspension was mounted on a glass slide covered with a cover slip and imaged immediately under fluorescence microscope.

3 Results

3.1 Purification of A. lakoocha agglutinin

Table 1 shows the purification scheme of A. lakoocha agglutinin. The crude seed extract on precipitation with 40% ammonium sulphate gave a protein of 18.2-fold purification with 80% yield. This upon treatment with rivanol achieved purification of 80-fold with 64% yield. Since the hemagglutination activity of ammonium sulphate precipitated fraction and rivanol precipitated lectin was inhibited by melibiose, affinity chromatography on melibiose-agarose column was used for isolation of pure lectin (Fig. 1). The rivanol precipitated lectin was totally absorbed to the column, which was eluted with deionized water following the technique of affinity repulsion chromatography. The specific activity of the purified A. lakoocha agglutinin designated as ALA was 8,62,000 with 4,205-fold purification. The recovered activity of the purified protein was 51.2% and minimum concentration of the protein required for erythro-agglutination was 3.6 ng/ml.
https://static-content.springer.com/image/art%3A10.1007%2Fs10719-008-9134-8/MediaObjects/10719_2008_9134_Fig1_HTML.gif
Fig. 1

Purification of Artocarpus lakoocha agglutinin (ALA). Affinity repulsion chromatographic profile of rivanol precipitated ALA on melibiose-agarose column (20 cm × 1 cm). Protein elution was monitored spectrophotometrically at 280 nm, and activity was observed by hemagglutination of normal human B erythrocytes

Table 1

Purification scheme of A. lakoocha lectina

Lectin Fraction

Total volume (ml)

Protein concen-tration (mg/ml)

Total protein (mg)

Total activity (HU)b

Specific activity (HU/mg)

Purification fold

Yield (%)

Crude extract

500

5.0

2500

512 × 103

205

1.0

100

Ammonium sulphate precipitated fraction

50

2.2

110

410 × 103

3,730

18.2

80.0

Rivanol precipitated lectin

20

1.0

20

328 × 103

16,400

80.0

64.0

Affinity purified lectin

2

0.152

0.304

262 × 103

862,000

4,205

51.2

aData shown are mean of three experiments

bHemagglutination unit (HU) is defined as minimum amount of protein (µg/ml) showing hemagglutination with normal human B erythrocytes

The homogeneity of ALA was proved by polyacrylamide gel electrophoresis as it produced a single band in non-denaturing acidic gel (Fig. 2a). However, ALA by SDS-PAGE (12%) under denaturing condition with or without 2-ME, produced two bands at ∼16 and ∼12 kD respectively indicating that ALA was a heterodimeric lectin (Fig. 2b). By gel filtration chromatography on Superose 6 column ALA gave a single symmetrical peak eluted at 41.9 ml corresponding to molecular mass ∼28 kD estimated by comparison with known protein standards (Fig. 2c). The absolute mass of the lectin was obtained as 27.3 kD from ESI-MS-Q-ToF mass analysis (Figure not shown).
https://static-content.springer.com/image/art%3A10.1007%2Fs10719-008-9134-8/MediaObjects/10719_2008_9134_Fig2_HTML.gif
Fig. 2

a Non-denaturing polyacrylamide gel (10%) electrophoresis of ALA. i Crude extract (10 μg), ii Purified ALA (10 μg). b SDS-PAGE of ALA on 12% polyacrylamide gel i Denatured and non-reduced ALA (10 μg), ii Precision Plus Protein™ Standards from Bio-Rad, iii Denatured and reduced ALA (10 μg). ALA was denatured by 1% SDS and reduced by 2% 2-mercaptoethanol at 100°C for 5 min. The protein bands were stained with Coomassie brilliant blue G-250. c Gel filtration chromatography of ALA on FPLC by Superose 6, 10/300 GL column; the elution was made by 20 mM TBS at 4°C at a flow rate of 24 ml/h and monitored at 280 nm

3.2 Amino acid sequencing and glycan detection

The N-terminal amino acid sequence of the 16 kD protein band of ALA and its comparison with that of other lectins of the same family (Moraceae) is presented in Table 2. The N-terminal sequence of 30 amino acid residues of ALA showed close resemblance with Maclura pomifera agglutinin (MPA), Artocarpus hirsuta lectin (AHL) and Jack fruit lectin, jacalin, all belonging to the same family, Moraceae. They have the sequence homology with ALA at 15, 19–21, 24–27 and 29 residues.
Table 2

Comparison of the N terminal sequence of ALA (16 kD) with other lectin members of the same family

 

 

1

 

 

 

 

 

 

 

 

10

 

 

 

 

 

ALA

 

A

S

Q

T

I

T

V

G

P

W

G

G

P

G

 

G

N

G

W

D

D

G

S

Y

T

G

I

R

Q

I

E

 

15

              

30

MPA

G

V

T

F

D

D

G

A

Y

T

G

I

R

E

I

 

AHL

G

K

A

F

D

D

G

A

F

T

G

I

R

E

I

 

JFL

G

K

A

F

D

D

G

A

F

T

G

I

R

E

I

 
Glycosylation pattern of the 16 kD subunit of ALA showed this subunit contained N-glycosidically linked “high mannose” or “hybrid”-type carbohydrate chains. 12 kD subunit did not react with the reagent indicating that this protein band might not be glycosylated (Fig. 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs10719-008-9134-8/MediaObjects/10719_2008_9134_Fig3_HTML.gif
Fig. 3

Glycan detection in ALA subunit. a NCM showing specificity of the lectins present in the kit. b NCM demonstrating the reaction of ALA with the lectins in the kit. Positive reaction obtained with GNA for 16 kD band. GNA (Galanthus nivalis agglutinin), SNA (Sambucus nigra agglutinin), MAA (Maackia amurensis agglutinin), PNA (Peanut agglutinin), DSA (Datura stramonium agglutinin)

3.3 SPR analysis

SPR studies on the interaction of ALA with different glycoproteins and glycans showed changes in RUs (Fig. 4) and the association constants of ligands with ALA are given in Table 3. Among the glycoproteins tested for binding assay asialo BSM showed highest binding which was 340 times higher than BSM. There was not much difference in intensities of binding of asialo glycophorin and Tn glycophorin with ALA. Asialo fetuin showed moderate binding with ALA, which being 57 times higher than fetuin was almost analogous to asialo glycophorin. Among the glycans, Tn containing glycopeptide showed strongest binding followed by GlcNAc Tn. Methyl-α-GalNAc was found two times higher in binding potency than GalNAc and 260 times higher than Methyl-α-Gal. T disaccharide, Galβ1-3GalNAc showed least binding with ALA among the glycans tested except Methyl-α-Gal.
https://static-content.springer.com/image/art%3A10.1007%2Fs10719-008-9134-8/MediaObjects/10719_2008_9134_Fig4_HTML.gif
Fig. 4

Sensorgram of the interactions of immobilized ALA with asialo BSM and T-disaccharide by SPR. The surface of the CM5 sensor chip was activated with N-hydroxysuccinimide (NHS) and N-ethyl-N″-3(diethylaminopropyl) carbodimide (EDC; 1:1) at a flow rate of 5 μl/min for 30 min. ALA (200 μg/ml) in 10 mM Na-acetate buffer (pH 4.8) was immobilized onto the chip and the blocking was performed with 1.0 M ethanolamine hydrochloride (pH 8.5). The reference flow cell was prepared in an analogous manner without ALA. Various concentrations (5, 10 and 20 nM) of asialo BSM and T-disaccharide were injected onto ALA- immobilized sensor chip at a flow rate 5 μl/min for 8 min

Table 3

Association constants for the binding of different glycoproteins and glycans to immobilized A. lakoocha agglutinin by SPR analysis

Glycan

Ka (M−1)

Asialo BSM

2.59 × 108

BSM

7.54 × 105

Asialo glycophorin

6.01 × 106

Tn glycophorin

5.37 × 105

Asialo fetuin

9.66 × 106

Fetuin

1.67 × 105

Galβ1, 3GalNAc

1.47 × 105

GlcNAc Tn

8.68 × 105

GalNAc

2.76 × 105

Tn containing glycopeptide

9.12 × 105

Me α GalNAc

6.10 × 105

Me α Gal

2.39 × 103

3.4 Biological properties

3.4.1 Mitogenic and antiproliferative activity

ALA was found to induce proliferation of human PBMC. The maximum proliferation was observed at a dose of 1 ng/ml of lectin (Fig. 5a). Thus, ALA could be designated as a very potent mitogen for human PBMC.
https://static-content.springer.com/image/art%3A10.1007%2Fs10719-008-9134-8/MediaObjects/10719_2008_9134_Fig5_HTML.gif
Fig. 5

a Mitogenic activity of ALA toward human PBMC observed as the count per min (CPM) of [H]3 thymidine uptake. bIn vitro inhibitory effect of ALA on proliferation of leukemic cell lines. IC50 on Jurkat cells, 13.15 ng/ml; IC50 on U937 cells, 25.14 ng/ml and K562 cells did not reach 50% inhibition even at 40 ng/ml

The antiproliferative activity of ALA on Jurkat cells was more intense (IC50 13.15 ng/ml) than its effect on U937 cells (IC50 25.14 ng/ml). K562 cells were less responsive than Jurkat and U937 cells since it did not reach 50% inhibition even at 40 ng/ml (Fig. 5b).

3.4.2 Leukemic cell identification by QD-ALA conjugate

Cadmium sulphide quantum dot-conjugated ALA was applied as an efficient fluorescent probe to distinguish between leukemic and normal cells. The qualitative schematic representation is given in Fig. 6. In the present study QD-tagged ALA was synthesized in such a way that the incipient nanoparticles of cadmium sulphide were stabilized by ALA with its potential hydrogen bond forming functionalities and the efficient interactions with thiol (–SH) groups of cystein residues of the peptide fragment of lectin. The most effective QD-ALA conjugate was formed when Cd+2 and ALA were present in 1:2 molar ratio (378.12 μg ALA per μg of CdS nanoparticle) in the solution from which the QD-ALA nanoconjugates were generated. This molar ratio was maintained throughout this study. The formation of ALA stabilized QDs was monitored by photoluminescence spectra of the reaction mixture in which an excitation at 350 nm produced the characteristic photoluminescence of lectin capped CdS nanoparticle (Fig. 7a). The average size of the QD-ALA adduct was calculated from the TEM image (Fig. 7b) and was on the average 10 nm in diameter. EDX, obtained from the TEM experiment, proved the encapsulation of CdS nanoparticles in the lectin molecule (Fig. 7c). A color luminescence image that was obtained from the original water soluble QD-ALA conjugate was taken as guideline of the emission color of the QD fluorophore (Fig. 7d).
https://static-content.springer.com/image/art%3A10.1007%2Fs10719-008-9134-8/MediaObjects/10719_2008_9134_Fig6_HTML.gif
Fig. 6

Qualitative schematic presentation of QD-ALA conjugate formation and its successive binding with leukemic cells. CdS nanocrystals are generated in situ by the reaction of CdCl2 and Na2S in presence of ALA, which contains T/Tn antigen binding sites. These fluorescent QD-ALA conjugates on incubation with leukemic cell lines having T/Tn antigenic determinants exposed on their cell surface, binds to the cells which are then imaged under fluorescence microscope

https://static-content.springer.com/image/art%3A10.1007%2Fs10719-008-9134-8/MediaObjects/10719_2008_9134_Fig7_HTML.gif
Fig. 7

a Photoluminescence spectra of QD-ALA conjugate. b TEM image of QD-ALA conjugate. c Electron dispersive X ray of QD. d Color luminescence images of QD-ALA

QD-ALA showed almost the same activity (titer 215 = 32,768) as the purified ALA (titer 217 = 131,072). The fluorescent microscopic images revealed that QD-ALA conjugate could successfully and specifically bind to leukemic cell lines, which appeared as green fluorescent dots (Fig. 8a,b,c) whereas there was no such dot with normal lymphocytes (Fig. 8d). The absence of green fluorescence in case of normal lymphocyte is a direct proof of the target specific interaction of the QD-ALA with leukemic cells. It was also clear from the fluorescent images that the green fluorescence of the QD-ALA conjugate was localized only on the surface of the leukemic cells.
https://static-content.springer.com/image/art%3A10.1007%2Fs10719-008-9134-8/MediaObjects/10719_2008_9134_Fig8_HTML.gif
Fig. 8

Fluorescent and phase contrast images of QD-ALA bound to a U 937; b Jurkat; c K562 cell lines; d Normal lymphocytes incubated with QD-ALA did not fluoresce

4 Discussion

A. lakoocha agglutinin purified by successive precipitation with ammonium sulphate and rivanol followed by affinity chromatography on melibiose-agarose column is a heterodimer comprising of two subunits 16 and 12 kD respectively and of molecular mass ∼28 kD. Heterodimeric nature of subunits of molecular weight 19 and 22 kD respectively was observed in Artocarpus altilis lectin, a member of Moraceae family [30]. The molecular weight of jacalin was reported to be 62 kD consisting of two non identical subunits of molecular weights 13 and 18 kD respectively [31]. Like jacalin, AHL and MPA, ALA showed strongest affinity for binding to Me-α-Gal/GalNAc.

The N-terminal sequence of ALA showed close homology to jacalin, AHL, MPA all belonging to Moraceae family. Thus ALA can easily be called a jacalin like lectin. Structural pattern of glycoform of ALA indicated mannose terminally linked to the 16 kD subunit and that the 12 kD subunit was non-glycosylated.

In the SPR studies ALA showed high binding affinity with glycoproteins in the order of asialo BSM > asialo fetuin ≅ asialo glycophorin > Tn glycopeptide ≅ Tn glycophorin ≅ GlcNAc Tn ≅ Me-α-GalNAc. BSM contains over 53% sialyl Tn and over 22% GlcNAc Tn (GlcNAcβ1→3Tn) as major carbohydrate side chain. The next abundant lectin determinant (lectin binding saccharide) in BSM is sialyl Tα and GlcNAc β1,6Tα. Complete removal of sialic acid by neuraminidase resulted in high abundance of Tn structure (Fig. 9) [32]. Glycophorin which is the human erythrocyte glycoprotein contains fifteen O-linked sialyl Tα structure. Disialylation of glycophorin resulted in asialoglycophorin which upon ß-galactosidase treatment produces Tn glycophorin (Fig. 9) [33].
https://static-content.springer.com/image/art%3A10.1007%2Fs10719-008-9134-8/MediaObjects/10719_2008_9134_Fig9_HTML.gif
Fig. 9

Glycan structures of mammalian glycoproteins

Fetuin contains six glycan chains per molecule; three of them are O-linked sialyl Tα attached to Ser or Thr residues of the protein core and the other three are N-linked sialyl bi- or tri antennary glycans of ratio 2:1. Thus asialo fetuin (Fig. 9) contains two lectin determinants Tα and II, a precursor of blood group antigens present at the non-reducing end of the carbohydrate chains of N-glycans [34]. ALA binds specifically to tumor associated carbohydrate antigens, GalNAcα1→Ser/Thr(Tn) and Galβ1→3GalNAcα1→Ser/Thr (Tα). SPR analysis confirmed our previous results of ELLSA which showed the binding affinity of ALA for Tn/Tα containing glycans in the order of asialo HSM > asialo OSM=asialo BSM > asialo PSM ≅ Tn glycophorin > asialo glycophorin ≅ Tn glycopeptides > monomeric Tn > GalNAc [15]. Although the binding studies of ALA with Tn/Tα glycotopes have been performed by different methods yet the binding order is almost the same. The enhanced specificity of ALA for Tn/Tα cluster suggests the importance of glycotopes polyvalency during carbohydrate-receptor interaction. Thus, ALA is considered a valuable reagent for analysis of progression of cancer since there is a direct link between carcinoma aggressiveness and the density of these antigens, including extent of tissue spread and vessel invasion [35, 36].

ALA was endowed with a potent mitogenicity effect toward human PBMC. It also exhibited a potent inhibitory effect on the proliferation of leukemic cell lines. This finding is in concurrence with the previous reports on many other lectins [28, 3740]. Lectins have been used for gastrointestinal cell targeting. Tomato lectin and jacalin retained their antitumor activity through the gastrointestinal tract [41].

The novelty and importance of newly developed CdS quantum dot-ALA nanoconjugate lies in differentiating between leukemic cells and normal lymphocytes since the lectin binding sites are cryptic in normal cells but are exposed in leukemic cells. The application of this nanoconjugate is based on the ability of its lectin part to recognize the specific target glycotopes on the cancer cell surface and to visualize this binding by novel fluorescence property of the QD. Introduction of CdS (QD) in the lectin based nanoconjugate also enhances its efficiency as compared to the conventional organic fluorescent dye tagged lectin sensors. Due to its very high surface/charge ratio more than one lectin molecule can interact with a single nanoparticle. Thus the nanoconjugate with multi-lectin molecule becomes more efficient in interacting with the cell surface glycoproteins rather than a conventional one to one lectin molecule-fluorescent dye conjugates. No green fluorescence was observed in normal lymphocytes. The TEM result justified the conjugation of ALA molecule with a single quantum dot nanocrystal, rather than the conjugation of one or more quantum dots with quaternary structure of ALA. This completely nullified any potential effect of QD on biological activity of ALA. Retention of hemagglutination activity of QD-ALA conjugate demonstrated it clearly. In this study the technique used was based on in situ generation of QD in presence of lectin leading to the formation of lectin capped QD to minimize the lectin-lectin cross linking.

In conclusion it can be stated that QD-ALA nanoconjugate is an efficient fluorescent marker for identification of several leukemic cell lines that gives rise to high quality images and possesses higher stability against photobleaching.

Acknowledgements

We sincerely thank Dr. Sujata Sharma of Department of Biophysics, All India Institutes of Medical Sciences, New Delhi for performing amino acid sequence analysis. The assistance of Dr. Syamal Roy of Indian Institute of Chemical Biology, Kolkata and Mr. Gautam Mondal of West Bengal University of Technology, Kolkata is gratefully acknowledged. This study was financially supported by a research grant (BT/PR4462/BRB/10/1350/2003) of the Department of Biotechnology, Government of India, New Delhi to B.P.C.

Copyright information

© Springer Science+Business Media, LLC 2008