Strong and specific interaction of ultra small superparamagnetic iron oxide nanoparticles and human activated platelets mediated by fucoidan coating

Abstract

Activated platelets play a pivotal role in cardiovascular diseases such as atherothrombosis. Therefore, strategies enabling activated platelet molecular imaging are of great interest. Herein, a chemical protocol was investigated for coating superparamagnetic iron oxide nanoparticles with low molecular weight fucoidan, a ligand of P-selectin expressed on the surface of activated platelets. The physico–chemical characterization of the obtained product demonstrated successful fucoidan coating and its potential as a T2 MRI contrast agent. The specificity and the strength of the interaction between fucoidan-coated iron oxide nanoparticles and human activated platelets was demonstrated by flow cytometry. Micromagnetophoresis experiments revealed that platelets experience magnetically-induced motion in the presence of a magnetic field gradient created by a micromagnet. Altogether, these results indicate that superparamagnetic iron oxide nanoparticles coated with low molecular weight fucoidan may represent a promising molecular imaging tool for activated platelets in human diseases.

---

Introduction

For the last twenty years, functionalized superparamagnetic iron oxide (SPIO) nanoparticles have been investigated as nano-structured smart biomaterials.^1–3 Easy ways of synthesis, the versatility of their surface modification and the possibility to develop multivalent systems made SPIO an eclectic tool for biomedical applications including magnetic resonance imaging contrast enhancers,^4–6 drug delivery,⁷ hyperthermia treatment of cancer,⁸ protein and cell separation,^9,10 biosensors¹¹ and tissue repair.¹² The biospecificity is achieved by conjugation of nanoparticles with small molecules,^13,14 polymers and oligomers,^15,16 antibodies, other proteins and peptides,^17,18 polysaccharides^19,20 or mixed compounds.²¹ More complex organic or inorganic core–shell structures have also been prepared to develop multimodal platforms for molecular imaging.^22,23

Platelets play a pivotal role in cardiovascular diseases such as atherothrombosis. When the atherosclerotic vessel is ruptured, thrombogenic substances are exposed and platelets contribute to the pathophysiologic process via adhesion, activation and aggregation stages.²⁴ Considering the importance of activated platelets in atherothrombosis, there is a great interest in approaches enabling activated platelet molecular imaging. In this context, P-selectin, a glycoprotein, expressed on the surface of activated platelets and on pathological vascular endothelium, is largely described as a candidate target for the diagnosis of vascular diseases such as atherosclerosis.^25,26 Fucoidan, heparin and dextran sulfate are sulfated polysaccharides which bind to P-selectin^27–30 mimicking the interaction with its main ligand P-selectin glycoprotein ligand-1 (PSGL-1) expressed on the membrane of leucocytes. Low molecular weight fucoidan was formerly demonstrated to be of potential interest for revascularization in cardiovascular diseases^31–33 and more recently found as the most efficient glycosidic ligand of P-selectin in purified system and in human whole blood experiments.²⁷ The large amount of P-selectin in platelet-rich thrombi formed after atherosclerotic plaque rupture or erosion makes it a good target for fucoidan. We recently demonstrated that ^99mTc radiolabelled fucoidan was able to detect different invascular thrombi and activated endothelium in different experimental models with a high sensitivity and specificity.³⁴

The aim of the present study was to combine the improved properties of SPIO and fucoidan in order to design a MRI molecular imaging tool able to interact with activated platelets in a strong and specific way. As a first step, a chemical protocol was investigated to conjugate fucoidan to ultra small SPIO. Next, the physico–chemical characterization of the obtained product was carried out in order to evaluate fucoidan coating and its potential as a T2 MRI contrast agent. Then, flow cytometry and micromagnetophoresis experiments were performed to evaluate the specificity and the strength of the interaction between activated platelets and fucoidan-conjugated SPIO.

Experimental

Starting materials

All chemicals, purchased from Sigma-Aldrich (St Quentin Fallavier, France), were of reagent grade and deionized water was used. Low molecular weight fucoidan from the brown seaweed Ascophyllum nodosum was commercialized by Algues & Mer (Ile d'Ouessant, France).

Starting USPIO

Superparamagnetic Fe₃O₄ (magnetite) nanocrystals were prepared by alkaline coprecipitation of FeCl₂ and FeCl₃ salts ([Fe] = 1.05 M).³⁵ Maghemite nanocrystals (γFe₂O₃) were then obtained by oxidizing magnetite (1.3 mol) into 1 L of nitric acid (2 N) containing 1.3 mol of iron nitrate under boiling.³⁶ After magnetic decantation and several washes, maghemite particles were dispersed in water at acidic pH.

Low molecular weight fucoidan

Chemical composition (w/w): fucose, 44.7 ± 5.4%; uronic acids, 22.8 ± 5.0%; sulfate groups, 17.6 ± 2.5%. Molecular weights: = 5000 g mol⁻¹; = 10 000 g mol⁻¹.

Synthesis of fucoidan coated-ultra small superparamagnetic iron oxide nanoparticules

Fucoidan amination. Fucoidan was aminated with diaminopropane at its reducing end. The protocol has been optimized based on methods described by Kondo et al. and Seo et al.^37,38 Briefly, 5.4 mL of 1.5 M diaminopropane in glacial acetic acid was added to 500 mg of fucoidan in a glass tube. The tube was sealed and heated to 90 °C for three hours. Then, the tube was cooled before adding 1.4 mL of 3 M dimethylborane in acetic acid. The tube was again sealed, heated to 90 °C for three more hours, cooled and then neutralized with diluted sodium hydroxide (NaOH). The product was then dialyzed (cut-off 1000 Da) five times against a carbonate buffer pH 9.6 with 1 M NaCl, five times against a water/ethanol (80/20) solution with 0.5 M NaCl and five times against distilled water. Finally, the product was frozen at −80°C and lyophilized.

The amount of primary amine was determined with phtalaldehyde colorimetric assay using bromopropylamine as the standard.³⁹

Chemical composition (w/w): fucose, 39.4 ± 5.0%; uronic acids, 13.8 ± 1.5%; sulfate groups, 29.7 ± 2.0%. Molecular weights: = 7800 g mol⁻¹; = 10 600 g mol⁻¹.

CMD coating of USPIO. In a second step, the starting ferrofluid was coated with carboxymethyldextran (CMD, = 15 000 g mol⁻¹, [COOH] = 1.3 mmol g⁻¹, Sigma-Aldrich, St Quentin Fallavier, France) following the method of Roger et al.³⁶ Briefly, 20 mL of starting ferrofluid was prepared at [Fe] = 5 × 10⁻² mol L⁻¹. The floculation step in acid medium was performed by the addition of CMD to the ferrofluid with [CMD]_final/[Fe]_final = 1%. The mixture was stirred for one hour and then centrifuged for one hour at 11 000 rev per min, the supernatant was removed and water was added to the pellet to the initial volume (20 mL). Then the re-suspension step in basic medium was performed by adding TMAH (tetramethylammonium hydroxide) 0.1 M to pH 10. The pH was maintained for one hour. The mixture was neutralized by adding 0.1 M HCl until a pH between 7.0 and 7.5, ultra-filtered on MicroSep 100 kDa (Pall, VWR France) and re-suspended in saline buffer (sterile water containing 0.15 M NaCl). Finally, the USPIO-CMD suspension was kept at 4 °C in the dark.

Aminated fucoidan coupling to USPIO-CMD. The CMD coated-USPIO was then coupled with aminated fucoidan by coupling reaction with conventional agents EDC/NHS (1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride/N-hydroxysuccinimide). Briefly, 5 mL of USPIO-CMD ([Fe] = 5 × 10⁻² mol L⁻¹) were stirred with 20 mg of EDC and 3.0 mg of NHS for 15 min at room temperature. Fucoidan previously aminated with diaminopropane at its reducing end was added to the mixture ([fucoidan] = 15 mg mL⁻¹) and maintained under agitation for 2 hours more. Purification was performed by dialyzing the suspension against NaCl 1 M (2×) and bisdistilled water (5×) before ultrafiltration on MicroSep 100 kDa (Pall, VWR France) and re-suspension in saline solution (sterile water containing 0.15 M NaCl). Finally, the fucoidan-USPIO suspension was kept at 4 °C in the dark.

Physical and chemical characterization

Magnetization measurement. The magnetization curves of the maghemite USPIOs were determined using a homemade vibrating sample magnetometer (VSM) under an applied magnetic field up to 0.93 T = 9300 Gauss (Fig. S1†). From the shape of the magnetization versus field intensity curve M(H), the size distribution of the magnetic cores was obtained by convolving the first order Langevin's law of paramagnetism L₁(ξ) = M/Φm_S = coth(ξ) − 1/ξ with ξ = μ₀m_SπD³H/6k_BT (k_B is the Boltzmann constant and μ₀ the vacuum magnetic permeability) with a Log-normal probability law of median diameter D₀ and width σ, defined as the standard deviation of the distribution ln(D/D₀)⁴⁰ For the starting nanoparticles (acid ones) the parameters of the size distribution determined by the Langevin law were d₀ = 7.5 nm and σ = 0.3.

Iron content per nanoparticle. The number of iron atom per nanoparticle can be estimated using the molar volume of the maghemite 15.77 cm³. For a nanoparticle of 10 nm in diameter the volume of the particle is (5.2 10⁻¹⁹ cm³) corresponds to approximately 20 000 iron atoms. This value decreases with the diameter and falls to 6800 for a 7 nm diameter.

Iron titration. The total iron concentration (mol L⁻¹) was determined by atomic absorption spectroscopy (AAS) with a Perkin-Elmer Analyst 100 apparatus after degradation of the USPIO in boiling HCl (35%). The volume fraction of iron oxide was deduced from the molar mass (159.7 g mol⁻¹) and mass density (5.1 g cm⁻³) of γ-Fe₂O₃, that is numerically Φ_USPIO (% v/v) = 1.577 [Fe] (mol L⁻¹).

Transmission electron microscopy. TEM images were recorded on a JEOL 2011 microscope working at 200 kV equipped with a GATAN Orius camera (Digital micrograph software) (Fig. S2†).

Dynamic light scattering and electrophoresis. The hydrodynamic diameter of USPIO was obtained in NaCl 0.15 M using a Zetasizer Nano ZS (Malvern Instrument) operating at the scattering angle 174°. The collective diffusion coefficient D was determined from the second-order autocorrelation function of the scattered light. From the value of the coefficient, the hydrodynamic diameter of the particle was calculated according to the Stokes–Einstein relation, D_H = k_BT/(3η_SD), where k_B is the Boltzmann constant, T the temperature (T = 298 K) and η_S the solvent viscosity (η_S = 0.89 × 10⁻³ Pa s for water). The autocorrelation functions of the scattered light are interpreted by using the CONTIN fitting procedure. Note that the hydrodynamic sizes appear larger than those determined by TEM because (i) TEM image was recorded with dry sample, (ii) light scattering is sensitive to the largest particles of the distribution.

Electrophoretic mobilitity (μ) measurements were carried out on the same instrument and converted into zeta potential (ζ) using Smoluchowski's approximation. All the measurements were performed at 25 °C and the data were at least the average of triplicate values.

Relaxivity of contrast agents. UC and UCF relaxivity measurements were performed with different concentrations of nanoparticles in pure water at 37 °C with relaxometers at 20 MHz (0.47 T; Minispec PC-120, Bruker, Karlsruhe, Germany) and at 60 MHz (1.42 T; Minispec mq-60,Bruker, Karlsruhe, Germany). MRI was carried out using a 7 T MRI system (Bruker Pharmascan). Nanoparticles were dispersed in 0.15% agarose and then allowed to gel at 4 °C. Images were acquired in T2-weighted sequences. The following parameters were used: FOV of 4 × 4 cm; matrix of 256 × 256; TR of 2500 ms; TE of 11 ms; slice thickness of 1 mm; flip angle of 180°.

Fucose and uronic acid contents. Fucose and uronic acid content were obtained by classical colorimetric assays of Dische and Schettles,⁴¹ and Bitter and Muir⁴² respectively.

Sulfur content. The sulfate content was obtained by formation of methylene blue after acidic hydrolysis of the samples, reduction of sulfate as hydrogen sulphide, and formation of methylene blue from N,N-dimethyl phenylene diamine dihydrochloride in strong acidic medium in presence of ferric chloride. Conditions were optimized from Gustaffson's work⁴³ and Kuban's work.⁴⁴

Briefly, 100 to 200 μL of the suspension of USPIO were added to 5 mL a reducing mixture prepared with 100 mL of concentrated hydriodic acid, 25 mL of glacial acetic acid and 2.5 g of hypophosphorous acid. The mixture was refluxed for 20 min through a water-cooled condenser under a stream of N₂ (100 mL min⁻¹) which carried away evolved hydrogen sulphide (H₂S). After bubling through a gas-washing column (20 mL of tris buffer 0.1 M, pH 7.2), H₂S was trapped as zinc sulphide in 30 mL of a solution of zinc acetate prepared by diluting 5 mL of 0.50 M and sodium acetate 0.10 M with 25 mL of deionized water. Eight mL of 16 mM ferric chloride in H₂SO₄ 0.1 M, and 2 mL of 3.7 mM N,N-dimethyl phenylene diamine dihydrochloride in H₂SO₄ 9 M were added to the zinc sulphide solution, and the final volume was adjusted to 50 mL with deionized water. The vial was maintained at room temperature under dark for 20 min and the absorption was measured at 665 nm with a UV-visible spectrophotometer (mc2, Safas, Monaco). The amount of sulfur was determined from a standard curve obtained with potassium sulfate solutions submitted to the overall process. This method did not require a special treatment of the samples and was not sensitive to ferric ions on the contrary to sulfur elemental analysis or turbidimetric assays.

In vitro evaluation of fucoidan-USPIO interaction with human platelets

Cytometry experiment-anti-P-selectin antibody competition. The interactions between coated USPIO and platelets were studied by flow cytometry as previously described.²⁷ Briefly, the blood of healthy adult donors was collected into tubes containing sodium citrate 3.8% (w/v). Citrated whole blood (5 μL) was diluted to one tenth in phosphate buffer saline (PBS) 1× at pH 7.4. Platelets were activated by the addition of TRAP (thrombin receptor activating peptide) to a final concentration of 20 μM. Five μL of anti-CD41 antibody PC5 (Phycoerythrin-Cyanine 5.1)-labeled (BD Biosciences, Le Pont de Claix France) were added to identify platelets. To study the effect of coated USPIO on the binding of anti-P-selectin on activated platelets, 5 μL of unlabeled coated USPIO ([Fe] = 0.02 M) and 5 μL of FITC (fluorescein isothiocyanate)-labelled anti-P-selectin antibody (BD Biosciences, Le Pont de Claix France) were added to the mixture. The samples were incubated at room temperature for 20 min. Then they were diluted with 1 mL of PBS before analysis by flow cytometry (LSRII cytometer, BD Biosciences, France). Ten thousand events had been collected for each sample.

Cytometry experiment-direct platelet labelling in whole blood. Blood from healthy adult donors was collected into tubes containing sodium citrate 3.8% (w/v). Citrated whole blood (5 μL) was diluted to one tenth in phosphate buffer saline (PBS) 1×. Platelets from blood samples were either activated or not by the addition of TRAP to a final concentration of 20 μM. Platelet population was gated by characteristic forward and side scatter and further identified by anti-CD41 antibody (Fig. S3†). Five μL of anti-CD41 conjugated to R Phycoerythrin (PE) (Beckman Coulter, Marseille France) were added to identify platelets. In order to investigate the specificity of UC and UCF in the binding of platelets in whole blood, 5 μL of UC or UCF ([Fe] = 0.02 M) were added to the mixture. The samples were incubated at room temperature for 20 min. Afterwards they were diluted with 2 mL of PBS before analysis by flow cytometry using a LSR II flow cytometer (Becton Dickinson, Inc.).

Platelet preparation for magnetophoresis. Blood from five healthy adult donors was collected into anticoagulant citrate dextrose acid (ACDA)-containing tubes (6 vols blood for 1 vol. citrate 3.8%, Vacutainer system, Beckton Dickinson, San Jose, USA). One tube per donor was used to perform whole blood flow cytometry experiments. Platelets from each donor were isolated and washed according to Jandrot-Perrus et al.⁴⁵ The base composition of the washing buffer was (final concentrations): 3.6 mM citric acid, 0.5 mM glucose, 0.5 mM KCl, 0.1 mM MgCl₂, 10.3 mM NaCl and 2 mM CaCl₂ prepared from concentrated stock solutions kept at −20 °C. The pH was adjusted to 6.5 with 4 M NaOH before adjusting the volume with bidistilled water to 250 mL. Just before use, 175 mg of bovine serum albumin and 100 mL of apyrase (from 5 mg mL⁻¹ stock solution) were added for every 50 mL, and 1 mL mL⁻¹ of 0.1 mM prostaglandin E1 in 70% ethanol. The base composition of the reaction buffer was (final concentrations): 0.55 mM glucose, 0.1 mM KCl, 0.2 mM MgCl₂, 13.7 mM NaCl, Hepes 0.5 mM, 0.03 mM H₂PO₄, 12 mM.

NaHCO₃ and 2 mM CaCl₂ prepared from concentrated stock solutions kept at −20 °C. The pH was adjusted to 7.4 with 4 M NaOH before adjusting the volume with bidistilled water. A 150 mg aliquot of bovine serum albumin was added just before use.

Isolated platelet incubation with nanoparticles. Six milliliters aliquots of isolated platelets (4 × 10⁸ platelets per mL) were prepared in centrifugation tubes. Activation was obtained with 75 μM TRAP. After incubation for 10 min at 37 °C, suspensions were centrifugated at 1200 g for 10 min at 20 °C. After discarding the supernatants, the pellets were resuspended in 10 mL of PBS 1× at pH 7.4. One mL of the suspension of activated platelets was incubated with either UC or UCF at 5 mM (iron concentration) for 15 min at 37 °C before use for magnetophoresis.

Platelet magnetophoresis. A drop of 10 μL of USPIO-activated platelets suspension was deposited into a magnetophoresis chamber featuring a glass slide/coverslip to which a 50 μm diameter nickel rod was integrated (Fig. S4A†). The outer extremity of the microrod was in direct contact to a rectangular magnet which created a 0.2 T uniform magnetic field. At a 50 μm distance from the microrod tip, a magnetic field gradient grad B of 285 T/m was created (Fig. S4B†). The magnetophoretic mobility of single platelets was tracked in this zone. In brief, the velocity of single platelets moving towards the nickel rod was observed with oil immersed 60X objective from an optical microscope (DMIRB Leica) connected to a CCD camera and a computer. The x,y-position of each individual platelet through the stack of successive frames captured at 132 ms time intervals was computed by Image J software. Platelet magnetophoresis follow-up was performed in triplicate and a total of 30 platelets were single tracked each time. No platelet magnetophoretic velocity was observed for the UC conditions, while UCF-bound platelets presented high magnetophoretic mobility in the presence of a magnetic field gradient.

For the UCF condition, the magnetic moment m of the platelets was then inferred from the measured magnetophoretic velocity. Briefly, the magnetic force mgra→dB, is balanced by the viscous drag 3πηd_hydυ→, where η is the viscosity of the carrier liquid [η = η_(water) = 10⁻³ Pa s at 20 °C]; d_hyd is the platelet diameter and υ→ is the measured velocity.

Results and discussion

In this study, fucoidan coated ultra small SPIO (USPIO) have been prepared from an aqueous magnetic fluid also called ferrofluid. The starting maghemite based ferrofluid was synthesized by co-precipitation of iron(II) and iron(III) chloride in concentrated alkaline medium according to Massart's process.³⁵ The final acidic ferrofluid ([Fe] = 1.05 M) was obtained after washings in acetone, diethylether and water. The average diameter of the bare nanoparticles determined by magnetization measurements was 7.5 nm. A strategy has been developped to prepare nanoparticles functionalized with a low molecular weight fucoidan as summarized on the Scheme 1. Fucoidan was aminated to its reducing end with diaminopropane followed by a reduction with dimethylborane in overall yields about 65% (w/w). The nanoparticles were first coated with carboxymethyldextran (CMD). Then a mixture of EDC (1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride) and NHS (N-hydroxysuccinimide) was used to activate carboxylic groups of CMD allowing the formation of amide bonds with aminated fucoidan. The starting nanoparticles, and nanoparticles coated with CMD and fucoidan were named U, UC and UCF respectively. UC and UCF were obtained in quantitative yields from simple and easy to handle chemical steps. Table 1 gathers physicochemical characterizations. The amount of fucoidan per nanoparticle was established from 110 to 330 by considering 0.41 sulfur atom per iron atom and an average amount of iron atoms per nanoparticle between 6800 and 20 000. The hydrodynamic diameters estimated from dynamic light scattering increased from 24 nm for bare nanoparticles to 38 nm for UC and up to 50 for UFC. The fucoidan coating was assessed by the determination of the sulfur content from sulfate groups. Carboxylate and sulfate groups exposed by the outer shell of the coated nanoparticles are responsible for the negative zeta potentials. Nanoparticles could be maintained at 4 °C in NaCl 0.15 M as stable suspensions for more than 1 month (data not shown). Longitudinal r1 and transverse r2 relaxivities measured in pure water were about 15 Fe mM⁻¹s⁻¹ and 120 to 130 Fe mM⁻¹ s⁻¹, respectively. Such values suggest that UFC can act as both T1 and T2 contrast agents but seem to be more favourable as T2 contrast agents due to their much larger r2 value. The contrast properties of UCF dispersed in agarose phantoms were tested using a 7T MRI (Fig. 1). Local magnetic field created by UCF resulting in an acceleration of the relaxation of surrounding proton magnetization was detected by MRI in T2-weighted images. As it can be observed, hyposignal increased when increasing UCF concentration in phantoms, demonstrating in vitro the MRI contrast properties of UCF.

Scheme 1 Preparation of coated USPIO. (i) Glacial acetic acid, 1.5 M diaminopropane, 90 °C, 3 h/dimethyl borane 3 M, 60 °C, 3 h (ii) floculation of ferrofluid with CMD ([CMD]final/[Fe]final = 1%), centrifugation, TAMH pH 10 for 1 h, neutralization. (iii) Coupling in water with EDC/NHS chemistry.

Table 1 Physico–chemical characterizations of USPIO

	Average size in 0.15 M NaCl (nm)	Zeta potential in 0.15 M NaCl (mV)	Relaxivity r1		Relaxivity r2		Fucoidan content
			20 MHz, 37 °C (Fe mM^−1 s^−1)	60 MHz, 37 °C (Fe mM^−1 s^−1)	20 MHz, 37 °C (Fe mM^−1 s^−1)	60 MHz, 37 °C (Fe mM^−1 s^−1)	Mole S/mole Fe	Estimated fucoidan per nanoparticle
Starting USPIO (U)	24 ± 4	37.7 ± 2.2	Nd	Nd	Nd	Nd	Na	Na
CMD-coated USPIO (UC)	33.8 ± 3.6	−10.3 ± 3.4	38.8 ± 1.9	37.5 ± 1.8	123.0 ± 6.2	137.0 ± 6.9	Na	Na
Fucoidan decorated UC (UCF)	47.9 ± 4.5	−14.3 ± 1.8	15. 5 ± 0.7	15.2 ± 0.7	119.0 ± 5.9	137.4 ± 6.9	0.41 ± 0.01	110–330

---

Fig. 1 MRI scans of agarose phantoms containing UCF at different concentrations demonstrating the contrast properties of nanoparticles.

Flow cytometry experiments were achieved by incubating human citrated whole blood with UC and UCF in presence of anti-P-selectin antibody (anti CD62P) labelled with FITC (fluorescein isothiocyanate) with or without TRAP (thrombin receptor activating peptide) as a platelet activator. Platelets were gated on side and forward scatter and their positivity for a fluorolabeled specific platelet antibody (anti-CD41). Inhibition of the anti-CD62P binding to activated platelets was observed in the presence of UCF as indicated by a decrease in the mean fluorescence intensity as it was not the case with UC (Fig. 2). Note that no aggregation of the nanoparticles could be observed in whole blood for the duration of each experiment (less than one hour). We conclude that fucoidan linked to the nanoparticles allowed their binding to activated platelets. The inhibition of anti CD62P binding demonstrated the specificity of the interaction, confirming that P-selectin is involved in UCF binding to activated human platelets. Additionally, it indicated that the interaction was strong enough to prevent the antibody to bind to P-selectin. Fucoidan immobilized on the surface of nanoparticles effectively acted as a competitor of the anti-P-selectin antibody.

Fig. 2 Normalized intensity of the fluorescence of FITC labelled anti-P-selectin measured from human whole blood flow cytometry. Platelets were activated with 20 μM TRAP and anti-CD62P was added: without USPIO (left bar); with UC (middle bar), with UCF (right bar). Error bars correspond to the SD from five donors.

In addition to estimate platelet binding via the inhibition of anti CD62P interaction by nanoparticles, complementary flow cytometry experiments were performed to investigate direct UC and UCF labelling to activated and non-activated platelets in whole blood. As compared to UC, UCF labelled more efficiently platelets (Fig. 3A). In fact, UCF/UC platelet labelling ratio was 2. For activated platelets (Fig. 3B), the UCF/UC labelling ratio was ∼4, which demonstrates that the interaction with activated platelets was increased by the fucoidan coating. In our experimental set-up, we could also detect USPIO binding to non-platelet cells, regardless of their coating. This binding was the same with and without TRAP indicating that it was not affected by the activation status of platelets. Hence, while USPIO could bind to both platelets and non-platelet cells, fucoidan coating endowed USPIO with enhanced affinity for activated platelets, an observation in line with fucoidan interaction with P-selectin expressed on activated platelets.

Fig. 3 Histograms of cyanine 5 fluorescence intensity for platelets and other cells in whole blood in the absence (A) or in the presence (B) of a platelet activator as a function of incubation with CMD-USPIO (UC) or fucoidan-coated USPIO (UCF), compared with negative control (no nanoparticles). Cyanine 5-A mean stands for the mean fluorescence intensity of cyanine 5-USPIO.

In order to confirm UCF binding to platelets and gain insight in the strength and extent of this interaction, a micromagnetophoresis experiment was performed.⁴⁶ Magnetophoresis experiments are a valuable tool to evidence cell interaction with magnetic nanoparticles.^47,48 Herein, the magnetic-induced mobility of platelets in interaction with magnetic nanoparticles was investigated. On one hand, the differential attraction of UC versus UCF nanoparticles attached to platelets was clearly evidenced by the massive attraction of the UCF-bound platelets to the magnetic tip, while no platelets were observed to move toward the tip in UC condition (Fig. 4). As a second step, the tracking of the migrating UCF-bound platelets allowed deriving quantitative information related to platelet magnetization: the magnetophoretic velocity of 43.2 ± 5.7 μm s⁻¹ was converted into a magnetic moment of (2.8 ± 0.4)10⁻¹⁵ A m². Interestingly such magnetization corresponds to (7.1 ± 0.9)10⁴ nanoparticles per platelet, which is in line with a total covering of the platelets surface by UCF, and in the same order of magnitude of the number of P-selectin molecules expressed at the surface of activated platelets.⁴⁹

Fig. 4 Micromagnetophoresis experiment to probe platelet interaction with CMD-USPIO (A) and fucoidan-coated USPIO (B).

Magnetophoresis experiments are also an important tool to distinguish strong interactions from weak unspecific ones.⁵⁰ In fact, weak interactions will not provide important contribution to magnetophoretic motion since they are not strong enough to afford the counterbalance of hydrodynamic drag force opposing the motion. Alternatively, strong magnetic nanoparticle binding to cells actively contributes to magnetic force overcoming hydrodynamic force, resulting in a movement toward the micromagnet tip. Therefore, the magnetically-induced motion experienced by platelets in the presence of a magnetic field gradient is a further proof of the strong binding of UCF to platelets.

In previous studies, we have demonstrated that ^99mTc radiolabelled LMW fucoidan could be used as contrast media for the imaging of vascular pathologies implicating the expression of P-selectin.³⁴ In a related nuclear imaging approach, ^99mTc conjugated with synthetic arginine–glycine–aspartic acid (RGD) analogue was investigated for binding the glycoprotein IIb/IIIa abundantly expressed on activated platelets.⁵¹ It consists on the platelet fibrinogen receptor (αIIbβ3, CD41/CD61) whose activation is the final common pathway of platelet activation, regardless of the platelet-activating stimulus. Peptide-conjugated ^99mTc could be captured by the thrombus and emitted radioactivity enabled detection by scintigraphic imaging in a clinical trial.⁵¹

Compared with nuclear imaging techniques, MRI offers improved spatial resolutions and concomitant anatomical information. Therefore, strategies relying on MRI have been tested for targeting the activated platelet receptor glycoprotein IIb/IIIa. As an example, iron oxide microparticles conjugated to an antibody against the glycoprotein IIb/IIIa have been investigated in a non occlusive mural thrombosis model.⁵² Similarly, a paramagnetic gadolinium chelate conjugated to cyclic peptide containing the RGD motif has been proposed as a molecular imaging agent for activated platelets via its binding to glycoprotein IIb/IIIa in a murine thrombosis model.⁵³ However, it may be important to highlight that other integrin subtypes (e.g., αvβ3 present on smooth muscle cells) could potentially compete for RGD binding, reducing selectivity against activated platelets.⁵⁴ Alternatively, some authors have made efforts to develop MRI contrast agents for imaging P-selectin.^55,56 However, these agents were composed of either an anti P-selectin antibody, or synthetic mimics of sialyl Lewis X (SLe^X) which is implicated in the binding of P-selectin with PSGL-1.³⁰ Although effective, the cost of synthesis of these agents including raw material and the risk of immunogenicity of antibodies are important drawbacks. Herein, an alternative approach was proposed in order to target P-selectin based on LMW fucoidan, which is a polysaccharide expected to present low immunogenicity, low cost and high specificity.

Conclusions

LMW fucoidan conjugation to USPIO was successfully achieved by the chemical protocol investigated herein. The obtained product featured about 110–330 fucoidan molecules per SPIO nanoparticle, as determined by sulfur content. It presented longitudinal r1 and transverse r2 relaxivities measured in pure water of about 15 Fe mM⁻¹ s⁻¹ and 120 to 130 Fe mM⁻¹ s⁻¹, respectively. This attests to its potential as T2 contrast agent, which was further confirmed by MRI. The ability of fucoidan-conjugated USPIO to interact with platelets was demonstrated. As attested by flow cytometry, fucoidan-conjugated USPIO were able to interact with platelets via P-selectin, demonstrating the specificity of the interaction. Additionally, micromagnetophoresis experiments revealed that the interplay between activated platelets and fucoidan-conjugated USPIO was strong enough to circumvent hydrodynamic drag force and induce the motion of single platelets as a whole toward a micromagnet. Altogether, the obtained results suggest that fucoidan-conjugated USPIO may represent a promising molecular MRI diagnostic tool able to interact with activated platelets in a strong and specific way.

Acknowledgements

Guerbet company (Aulnay sous Bois, France) is gratefully acknowledged for relaxivity measurements. The authors acknowledge Institut Claude Bernard for MRI. A.K.A. Silva was supported by a grant from NanoAthero FP-7 project (Grant NMP-LA-2012-309820).

Notes and references

1. O. Veiseh, J. W. Gunn and M. Zhang, Adv. Drug Delivery Rev., 2010, 62, 284–304.
2. S. Laurent, D. Forge, M. Port, A. Roch, C. Robic, L. Vander Elst and R. N. Muller, Chem. Rev., 2008, 108, 2064–2110.
3. C. Corot, P. Robert, J. M. Idee and M. Port, Adv. Drug Delivery Rev., 2006, 58, 1471–1504.
4. U. I. Tromsdorf, O. T. Bruns, S. C. Salmen, U. Beisiegel and H. Weller, Nano Lett., 2009, 9, 4434–4440.
5. J. R. McCarthy and R. Weissleder, Adv. Drug Delivery Rev., 2008, 60, 1241–1251.
6. C. Corot, K. G. Petry, R. Trivedi, A. Saleh, C. Jonkmanns, J. F. Le Bas, E. Blezer, M. Rausch, B. Brochet, P. Foster-Gareau, D. Baleriaux, S. Gaillard and V. Dousset, Invest. Radiol., 2004, 39, 619–625.
7. N. Kohler, C. Sun, A. Fichtenholtz, J. Gunn, C. Fang and M. Q. Zhang, Small, 2006, 2, 785–792.
8. N. A. Brusentsov, V. V. Gogosov, T. N. Brusentsova, A. V. Sergeev, N. Y. Jurchenko, A. A. Kuznetsov, O. A. Kuznetsov and L. I. Shumakov, J. Magn. Magn. Mater., 2001, 225, 113–117.
9. H. Gu, K. Xu, C. Xu and B. Xu, Chem. Commun., 2006, 941–949.
10. I. Safarik and M. Safarikova, Biomagn. Res. Technol., 2004, 2, 7.
11. J. M. Perez, L. Josephson, T. O'Loughlin, D. Hogemann and R. Weissleder, Nat. Biotech., 2002, 20, 816–820.
12. A. Solanki, J. D. Kim and K. B. Lee, Nanomedicine, 2008, 3, 567–578.
13. Y. Lalatonne, C. Paris, J. M. Serfaty, P. Weinmann, M. Lecouvey and L. Motte, Chem. Commun., 2008, 2553–2555.
14. S. A. Corr, A. O'Byrne, Y. K. Gun'ko, S. Ghosh, D. F. Brougham, S. Mitchell, Y. Volkov and A. Prina-Mello, Chem. Commun., 2006, 4474–4476.
15. A. Masotti, A. Pitta, G. Ortaggi, M. Corti, C. Innocenti, A. Lascialfari, M. Marinone, P. Marzola, A. Daducci, A. Sbarbati, E. Micotti, F. Orsini, G. Poletti and C. Sangregorio, Magma, 2009, 22, 77–87.
16. A. Petri-Fink, M. Chastellain, L. Juillerat-Jeanneret, A. Ferrari and H. Hofmann, Biomaterials, 2005, 26, 2685–2694.
17. J. Xie, J. Wang, G. Niu, J. Huang, K. Chen, X. Li and X. Chen, Chem. Commun., 2010, 46, 433–435.
18. A. Tsourkas, V. R. Shinde-Patil, K. A. Kelly, P. Patel, A. Wolley, J. R. Allport and R. Weissleder, Bioconjugate Chem., 2005, 16, 576–581.
19. F. P. Gao, Y. Y. Cai, J. Zhou, X. X. Xie, W. W. Ouyang, Y. H. Zhang, X. F. Wang, X. D. Zhang, X. W. Wang, L. Y. Zhao and J. T. Tang, Nano Res., 2010, 3, 23–31.
20. R. S. Molday and D. MacKenzie, J. Immunol. Methods, 1982, 52, 353–367.
21. F. M. Kievit, O. Veiseh, N. Bhattarai, C. Fang, J. W. Gunn, D. Lee, R. G. Ellenbogen, J. M. Olson and M. Zhang, Adv. Funct. Mater., 2009, 19, 2244–2251.
22. J. F. Zhou, L. J. Meng, Q. H. Lu, J. W. Fu and X. B. Huang, Chem. Commun., 2009, 6370–6372.
23. A. H. Lu, E. L. Salabas and F. Schuth, Angew. Chem., Int. Ed. Engl., 2007, 46, 1222–1244.
24. L. K. Jennings, Am. J. Cardiol., 2009, 103, 4A–10A.
25. E. Canet-Soulas and D. Letourneur, Magma, 2007, 20, 129–142.
26. T. J. Molenaar, J. Twisk, S. A. de Haas, N. Peterse, B. J. Vogelaar, S. H. van Leeuwen, I. N. Michon, T. J. van Berkel, J. Kuiper and E. A. Biessen, Biochem. Pharmacol., 2003, 66, 859–866.
27. L. Bachelet, I. Bertholon, D. Lavigne, R. Vassy, M. Jandrot-Perrus, F. Chaubet and D. Letourneur, Biochim. Biophys. Acta, 2009, 1790, 141–146.
28. T. V. Pochechueva, O. E. Galanina, N. A. Ushakova, M. E. Preobrazhenskaya, M. A. Sablina, N. E. Nifantiev, Y. V. Tsvetkov, Y. V. Vozney, A. Imberty and N. V. Bovin, Glycoconjugate J., 2004, 20, 91–97.
29. A. Koenig, R. Jain, R. Vig, K. E. Norgard-Sumnicht, K. L. Matta and A. Varki, Glycobiology, 1997, 7, 79–93.
30. A. Varki, Proc. Natl. Acad. Sci. U. S. A., 1994, 91, 7390–7397.
31. A. C. Lake, R. Vassy, M. Di Benedetto, D. Lavigne, C. Le Visage, G. Y. Perret and D. Letourneur, J. Biol. Chem., 2006, 281, 37844–37852.
32. C. E. Luyt, A. Meddahi-Pelle, B. Ho-Tin-Noe, S. Colliec-Jouault, J. Guezennec, L. Louedec, H. Prats, M. P. Jacob, M. Osborne-Pellegrin, D. Letourneur and J. B. Michel, J. Pharmacol. Exp. Ther., 2003, 305, 24–30.
33. J. F. Deux, A. Meddahi-Pelle, A. F. Le Blanche, L. J. Feldman, S. Colliec-Jouault, F. Bree, F. Boudghene, J. B. Michel and D. Letourneur, Arterioscler., Thromb., Vasc. Biol., 2002, 22, 1604–1609.
34. F. Rouzet, L. Bachelet-Violette, J. M. Alsac, M. Suzuki, A. Meulemans, L. Louedec, A. Petiet, M. Jandrot-Perrus, F. Chaubet, J. B. Michel, D. Le Guludec and D. Letourneur, J. Nucl. Med., 2011, 52, 1433–1440.
35. R. Massart, IEEE Trans. Magn., 1981, 17, 1247–1248.
36. J. Roger, J. N. Pons, R. Massart, A. Halbreich and J. C. Bacri, Eur. Phys. J.: Appl. Phys., 1999, 5, 321–325.
37. A. Kondo, J. Suzuki, N. Kuraya, S. Hase, I. Kato and T. Ikenaka, Agric. Biol. Chem., 1990, 54, 2169–2170.
38. J. H. Seo, K. Adachi, B. K. Lee, D. G. Kang, Y. K. Kim, K. R. Kim, H. Y. Lee, T. Kawai and H. J. Cha, Bioconjugate Chem., 2007, 18, 2197–2201.
39. M. Roth, Anal. Chem., 1971, 43, 880–882.
40. R. W. Chantrell, J. Popplewell and S. W. Charles, IEEE Trans. Magn., 1978, 14, 975–977.
41. Z. Dische and L. B. Schettles, J. Biol. Chem., 1948, 175, 595.
42. T. Bitter and H. M. Muir, Anal. Chem., 1962, 4, 330–334.
43. L. Gustafsson, Talanta, 1960, 4, 227–235.
44. V. Kuban, P. K. Dasgupta and J. N. Marx, Anal. Chem., 1992, 64, 36–43.
45. M. Jandrot-Perrus, D. Didry, M. C. Guillin and A. T. Nurden, Eur. J. Biochem., 1988, 174, 359–367.
46. A. K. A. Silva, R. Di Corato, F. Gazeau, T. Pellegrino and C. Wilhelm, Nanomedicine, 2012, 7, 1713–1727.
47. C. Wilhelm, C. Billotey, J. Roger, J. Pons, J. C. Bacri and F. Gazeau, Biomaterials, 2003, 24, 1001–1011.
48. C. Wilhelm, F. Gazeau and J. Bacri, Eur. Biophys. J., 2002, 31, 118–125.
49. R. P. McEver, Thromb. Haemostasis, 2001, 86, 746–756.
50. D. Fayol, N. Luciani, L. Lartigue, F. Gazeau and C. Wilhelm, Adv. Healthcare Mater., 2013, 2, 312.
51. S. M. Bates, J. Lister-James, J. A. Julian, R. Taillefer, B. R. Moyer and J. S. Ginsberg, Arch. Intern. Med., 2003, 163, 452.
52. C. Von zur Muhlen, D. Von Elverfeldt, J. Moeller, R. Choudhury, D. Paul, C. Hagemeyer, M. Olschewski, A. Becker, I. Neudorfer and N. Bassler, Circulation, 2008, 118, 258–267.
53. A. Klink, E. Lancelot, S. Ballet, E. Vucic, J.-E. Fabre, W. Gonzalez, C. Medina, C. Corot, W. J. Mulder and Z. Mallat, Arterioscler., Thromb., Vasc. Biol., 2010, 30, 403–410.
54. I. Y. Chen and J. C. Wu, Circulation, 2011, 123, 425–443.
55. R. T. Haverslag, G. Pasterkamp and I. E. Hoefer, Cardiovasc. Hematol. Disord.: Drug Targets, 2008, 8, 252–260.
56. M. A. McAteer, J. E. Schneider, Z. A. Ali, N. Warrick, C. A. Bursill, C. von zur Muhlen, D. R. Greaves, S. Neubauer, K. M. Channon and R. P. Choudhury, Arterioscler., Thromb., Vasc. Biol., 2008, 28, 77–83.

---

Footnotes

† Electronic supplementary information (ESI) available: Supplementary figures. See DOI: 10.1039/c3ra46757k

‡ L. Bachelet-Violette and A. K. A. Silva have equally contributed to this study.

---