shear-regulated uptake of nanoparticles by endothelial cells and development of...

Shear-regulated uptake of nanoparticles by endothelialcells and development of endothelial-targetingnanoparticles

Arthur Lin,1,2* Abhimanyu Sabnis,1,2* Soujanya Kona,1,2 Sivaniaravindapriya Nattama,1,2

Hemang Patel,3 Jing-Fei Dong,4 Kytai T. Nguyen1,2,3

1Department of Bioengineering, University of Texas at Arlington, Arlington, Texas2Joint Biomedical Engineering Program, University of Texas at Southwestern Medical Center at Dallas, Dallas, Texas3Department of Irrigation and Biological Engineering, Utah State University, Logan, Utah4Department of Medicine, Baylor College of Medicine, Houston, Texas

Received 30 April 2008; revised 9 April 2009; accepted 3 June 2009Published online 3 August 2009 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.a.32592

Abstract: The purpose of this research project was to de-velop nanoparticles with improved targeting, adhesion,and cellular uptake to activated or inflamed endothelialcells (ECs) under physiological flow conditions. Our hy-pothesis is that by mimicking platelet binding to activatedECs through the interaction between platelet glycoproteinIba (GP Iba) and P-selectin on activated endothelial cells,GP Iba-conjugated nanoparticles could exhibit increasedtargeting and higher cellular uptake in injured or activatedendothelial cells under physiological flow conditions. Totest this hypothesis, fluorescent-carboxylated polystyrenenanoparticles were selected for the study as a model parti-cle because of its narrow size distribution as a ‘‘proof-of-concept.’’ Using confocal microscopy, fluorescent measure-ments, and protein assays, cellular uptake properties were

characterized for these polystyrene nanoparticles. Thestudy also found that conjugation of 100-nm polystyrenenanoparticles with glycocalicin (the extracellular segmentof GP Iba) significantly increased the particle adhesion onP-selectin-coated surfaces and cellular uptake of nanopar-ticles by activated endothelial cells under physiologicalflow conditions. The results demonstrate that these novelendothelial-targeting nanoparticles could be the first steptoward developing a targeted and sustained drug deliverysystem that can improve shear-regulated particle adhesionand cellular uptake. � 2009 Wiley Periodicals, Inc.J Biomed Mater Res 93A: 833–842, 2010

Key words: nanoparticles; endothelial cells; shear stress;adhesion; targeting

INTRODUCTION

Recent studies have focused on targeting endothe-lium with nanoparticles to deliver drug and/or genetherapy for a variety of pathological conditions in thevascular system because of its large population andproximity to blood flow.1–6 Under diseased conditionssuch as thrombosis, inflammation, and restenosis, en-

dothelial cells (ECs) become activated and express en-dothelial cell adhesion molecules (ECAMs) such asP-selectin and E-selectin.3–5,7,8 Using ECAM ligandshas been considered for effective targeting of diseasedECs. For instance, immunoliposomes and micropar-ticles coated with ligands (sLex, PSGL-1) and human-ized monoclonal antibodies (mAbs) bound to P-selec-tin highly expressed on activated ECs have demon-strated varying degrees of success.4,7–12

The long-term goal of our research is to developbiodegradable nanoparticles that can target anddeliver therapeutic agents to treat injured andinflamed ECs after angioplasty and/or stenting treat-ments. Using carboxylated polystyrene (nonbiode-gradable) nanoparticles as model particles, an effec-tive targeting strategy will be determined in thisstudy. These nanoparticles have a uniform composi-tion, narrow size distribution, and are stable. Theseproperties of the carboxylated polystyrene particleswill allow us to investigate the adhesion and uptake

*These authors contributed equally to this work.Correspondence to: K. T. Nguyen; e-mail: knguyen@

uta.eduContract grant sponsor: National Science Foundation;

contract grant number: 0215852Contract grant sponsor: American Heart Association Sci-

entist Development Award; contract grant number:0735270NContract grant sponsor: NIH; contract grant number:

HL091232

� 2009 Wiley Periodicals, Inc.

properties without concern for the possibility thatthe particle size and degradation influence on thetargeting strategy would affect the outcome. GP Ibawas selected as an ideal ligand for conjugating ontothe surface of nanoparticles to increase their adhe-siveness under high shear conditions. GP Iba is wellknown for its role on the platelet adhesion onto thevascular wall in the high shear stress regions.13,14 Itcan also serve as a targeting ligand that binds specif-ically to P-selectin expressed on activated ECs.13,15

These ‘‘platelet-mimicking nanoparticles’’ should spe-cifically adhere onto damaged or activated ECsunder conditions of high shear stress, inducing cellu-lar retention and uptake of nanoparticles (Fig. 1).

Understanding the cellular uptake process of poly-styrene nanoparticles and characterizing the interac-tion of platelet-mimicking polystyrene nanoparticleswith P-selectin and activated ECs is needed to reach along-term goal to develop drug-loaded endothelial-targeting nanoparticles. Although previous studiessuccessfully characterized the cellular uptake of poly-styrene nanoparticles in the human colon adenocarci-noma cell line (Caco-2) and umbilical vein endothelialcells (HUVECs),1,16,17 these cell types are differentfrom the human aortic endothelial cells (HAECs)used in this study. Since we want to target restenosisafter vascular intervention, HAECs are a better choicefor in vitro cell culture characterization. In this study,the cellular uptake properties of polystyrene nano-particles of various sizes, doses, and incubation timeswere investigated using HAECs. The effects of shearstress on the cellular uptake of 100-nm size nanopar-ticles by HAECs were also studied using the parallel-plate flow chamber system. The cellular uptake ofthese particles in HAECs was then determined usingboth fluorescent intensity measurements and confocalmicroscopy imaging of cells after the incubation time.In addition, purified glycocalicin (the extracellularsegment of platelet GP Iba) was conjugated to the

polystyrene nanoparticles to produce GP Ib nanopar-ticles. The experiments were then performed to deter-mine the effects of GP Ib conjugation on particle ad-hesion on the P-selectin-coated surfaces and cellularuptake of these nanoparticles in activated HAECsunder physiological flow conditions.

MATERIALS AND METHODS

Materials

Fluoresbrite1 YG Carboxylate Polystyrene particleswere purchased from Polysciences (Warrington, PA).Chemicals, if not specified, were purchased from Sigma-Aldrich (St. Louis, MO). HAECs and low serum growthsupplement [LSGS, 2% fetal bovine serum, hydrocortisone(1 lg/mL), human epidermal growth factor (10 ng/mL),basic fibroblast growth factor (3 ng/mL), and heparin(10 lg/mL)] were purchased from Cascade Biologics (Port-land, OR). Cell culture media, supplements, and buffers,including Medium 199 (M199), trypsin, fetal bovine serum(FBS), penicillin–streptomycin, and phosphate-buffered sa-line (PBS) were purchased from Invitrogen Corp.(Carlsbad, CA).

Cell culture

HAECs were grown in M199 supplemented with 1%penicillin–streptomycin and LSGS (complete M199). Uponreaching confluence, cells were passaged or used for experi-ments. Cells up to passage 9 were used for the study.

Cellular uptake of polystyrene nanoparticles byhuman aortic endothelial cells

To determine the cellular uptake of nanoparticles,HAECs were seeded onto 24-well plates at a density of30,000 cells/well and allowed to grow for 2 days. Sincehigh serum media would contribute to the variation in theassessment of particle uptake in HAECs,18 we chose to uselow serum media in all particle uptake studies. The firststudy was conducted to determine the effects of particlesize on cellular particle uptake in HAECs. Particles rang-ing in size from 100 to 1000 nm were suspended in low se-rum growth medium at the concentration of 100 lg/mL.Medium from the 24-well plate was replaced with the par-ticle suspensions, and cells were allowed to incubate withthe nanoparticles for 1 h. The second study was to opti-mize nanoparticle concentration. Nanoparticle (100 nm insize) solutions (100–800 lg/mL) were prepared in low se-rum growth medium and added to HAECs for 1 h. Thethird study was to evaluate the effect of incubation timeon cellular uptake of nanoparticles (100 nm in size). Cellswere incubated with 100 lg/mL nanoparticle solution forvarious time periods up to 6 h. Cells without nanoparticlesolutions served as controls.

An additional study was performed to investigate theeffects of shear stress on cellular uptake of nanoparticles

Figure 1. Platelet-mimicking nanoparticles for targetingdysfunctional endothelium. [Color figure can be viewed inthe online issue, which is available at www.interscience.wiley.com.]

834 LIN ET AL.

Journal of Biomedical Materials Research Part A

by HAECs. Cells were seeded onto pre-etched glass slidesat a density of 105 cells/cm2. Upon reaching confluence,cells on glass slides were exposed to either 1 or 5 dyne/cm2 of media containing 100-nm nanoparticles at a concen-tration of 100 lg/mL for 30 min using the parallel-plateflow chamber system as described previously.19 The paral-lel flow chamber was chosen over other flow systemsbecause of its ability to produce constant levels of shearstress. Cells in static condition served as the control.

Quantification of nanoparticle uptake by HAECsusing a fluorometer

After these experiments, cells were washed carefully atleast three times with cold PBS to remove any remainingnanoparticles. After washing, 1 mL of 1% Triton1 X-100was added to each cell sample and incubated for 1 h tolyse the cells. Twenty-five microliters of cell lysate fromeach well was used to determine the total cellular proteincontent using the BCA protein assay (Pierce) following themanufacturer’s instructions. Total protein concentration ineach sample was used for normalizing cellular uptake ofnanoparticles. To quantify nanoparticle uptake by HAECs,the fluorescent intensity in cell lysates was measured atEm 480 nm/Ex 510 nm (VersaFluorTM, Bio-Rad Laborato-ries, Hercules, CA). A standard curve was obtained by se-rial dilution of same stock nanoparticle solutions in 1%Triton1 X-100, and the fluorescent intensities were mea-sured using the same filters. The nanoparticle uptake byHAECs was calculated by normalizing the nanoparticleconcentration in each cell lysate sample with the total cel-lular protein, which correlates to the number of cells in thesample. For size uptake, we seeded HAECs at differentseeding densities and generated a standard plot of varyingnumber of known cells to a standard of known BSA. Wethen calculated the amount of protein per 1000 cells andconverted the cellular uptake per total cellular protein tothe cellular uptake per 1000 cells.

Evaluation of nanoparticle uptake usingconfocal microscopy

After these experiments, cover slips were washed withcold PBS, followed by the addition of cold FM1 4-64 FX(5 lg/mL of Texas-Red1 dye, Invitrogen) in PBS for 10 minto stain cell plasma membranes. HAECs were then imagedusing a confocal laser scanning microscope (Carl Zeiss LSMMeta 510, Goettingen, Germany) equipped with FITC andRITC filters [Ex(l) 488 nm, Em(l) 543 nm]. Images ofHAECs were taken using a fast scan option to section thecells. Slice thickness was set at 1 lm, with an average of 18slices taken per image. The images were then analyzedusing Carl Zeiss LSM Image Browser (version 3.5).

Nanoparticle conjugation with glycocalicin

Carboxylated nanoparticles, 100 nm in size, were conju-gated with glycocalicin using carbodiimide chemistry andavidin–biotin affinity. First, glycocalicin were biotinylated

using the Biotin-X-NHS Kit (EMD Biosciences, San Diego,CA) following the manufacturer’s instructions. Second,polystyrene nanoparticles were added to a 15 mg/mLEDC solution in 0.1M MES buffer, pH 4.75, and incubatedat room temperature for 4 h to ensure carboxyl group acti-vation. After 4 h, 500 lg of avidin (EMD Biosciences) wasadded to the nanoparticle solution and allowed to interactovernight in 0.1M sodium bicarbonate solution (pH 8.5).Biotinylated glycocalicin (120 lL of 50 lg/mL) describedin the first step was added to the avidin-conjugated poly-styrene nanoparticle solution and reacted at room tempera-ture under gentle agitation for 1 h. The nanoparticle solu-tion was then dialyzed against 0.1M PBS for 3 h to removeany unreacted materials. Care was taken to avoid exposureof the nanoparticles to light throughout the entire proce-dure. To confirm the conjugation of ligands onto nanopar-ticles, 100 lL of 30 lg/mL primary mouse antibody mono-clonal against glycocalicin (HIP1, BioLegend) was addedto glycocalicin-conjugated nanoparticles in PBS. A fluores-cent (red) secondary antibody (anti-mouse IgG1, BioLe-gend) was added to the nanoparticle solution and incu-bated for 1 h. After washing, nanoparticles were analyzedusing confocal microscopy method.

Preparation of P-selectin-coated slidesand activated ECs

To prepare P-selectin-coated surfaces, glass slides wereincubated with 500 lL of 20 lg/mL P-selectin (R&D Sys-tems) for 4 h at 378C, followed by 1 h of incubation with a1% BSA solution in PBS to block any nonspecific bindingsites. Half the slides were further incubated with P-selectinantibodies for 1 h at room temperature to serve as a nega-tive control. The slides were then washed gently with 0.9%NaCl solution to remove any unbound P-selectin or anti-body. To prepare activated HAECs, HAECs seeded onglass slides as described earlier were treated with 25 mMhistamine for 12 min at room temperature to induceP-selectin expression on HAECs. Stimulated (activated)cells were used immediately in flow chamber experiments.

Effects of GPIb conjugation on particle adhesionand cellular uptake under physiologicalflow conditions

Slides (coated with either P-selectin or P-selectin/anti-P-selectin for surface studies and containing HAEC mono-layer for cell studies) were assembled into the parallel-plate flow systems.19 For noncellular flow studies, the flowchamber was set to produce 5 dyne/cm2 of shear stress,and three types of samples were used. These samples werePBS solution consisting of control nanoparticles, GP Iba-conjugated nanoparticles, and GP Iba-conjugated nanopar-ticles preincubated with antibodies against GP Iba (or GPIba mAbs). After the flow experiments, the amount ofnanoparticles bound to the glass cover slides were mea-sured using a fluorometer. The glass slides were alsoobserved using confocal microscopy. For cell studies, theshear stress was varied between 0 and 15 dyne/cm2. Thenanoparticle solutions were diluted to 100 lg/mL concen-

SHEAR-REGULATED UPTAKE OF NANOPARTICLES 835


tration in low serum media for cell studies. ActivatedHAECs seeded on the glass slides were exposed to low se-rum media consisting of either nonconjugated (control) orGPIb-conjugated nanoparticles under various levels ofshear stress. Samples in the static condition served ascontrols.

Statistical analysis

Analysis of the results was performed using ANOVAand t-tests with p < 0.05 (StatView 5.0 software, SAS Insti-tute). Posthoc comparisons were made using the Fisher’sleast significant differences (LSDs). All the results aregiven as mean 6 SD.

RESULTS

Characteristics of cellular uptake of nanoparticlesby HAECs

To determine the effects of nanoparticle size oncellular uptake by HAECs, experiments were con-ducted using polystyrene nanoparticles of varyingsizes (100–1000 nm) due to their narrow size distri-bution. The results displayed a clear trend ofdecreased cellular uptake with increase in nanopar-ticle size. Particles of 1000 nm displayed least uptakeby HAECs, whereas the smallest nanoparticles(100 nm) were uptaken the most [Fig. 2(A)].

Nanoparticle uptake for 100-nm polystyrene nano-particles displayed trends toward to the concentra-tion and incubation time dependency. For theformer, cellular uptake by HAECs reached saturationat the nanoparticle concentration of 200 lg/mL[Fig. 2(B)]. For the latter, the saturation was reachedin 30 min for the polystyrene nanoparticles [Fig.2(C)]. Confocal microscopy was used to confirm theuptake of particles inside the cells (Fig. 3). Images atthe middle point of the cell confirmed that the fluo-rescent nanoparticles were localized inside cells.

To investigate the effects of shear stress on cellularuptake of nanoparticles by HAECs, the parallel-plateflow chamber system was used to generate differentlevels of shear stress. We found that the cellularuptake of nanoparticles was decreased with theincrease of shear stress magnitude (Fig. 4), indicatingan inverse correlation between nanoparticle uptakeand the levels of shear stress.

Conjugation of GPIb-enhanced adhesion ofparticles on P-selectin surfaces and cellular uptakeof particles by activated HAECs

Because the size of nanoparticles is 100 nm, belowthe detection limit of the available flow cytometry,

the conjugation of GP Ib onto nanoparticles wasdetermined using the Zeiss Cytoviva MicroscopeImaging System (at a magnification of 103). Asshown in Figure 5, fluorescent-tagged secondary

Figure 2. (A) Effects of the particle size on the cellularuptake of polystyrene nanoparticles in ECs. Values wereobtained after 1 h of incubation with nanoparticle solu-tions and represented as mean 6 SD (n 5 6). **Significantdifferences when compared with the 100-nm nanoparticlesamples (p < 0.001). (B) Effect of concentration on the cel-lular uptake of 100-nm polystyrene nanoparticles in ECs.Values were obtained after 1 h of incubation with nano-particle solutions and represented as mean 6 SD (n 5 6).(C) Effect of the incubation time on the cellular uptake of100-nm polystyrene nanoparticles in ECs. Samples wereincubated with 100 lg/mL of nanoparticle solution. Valuesrepresent mean 6 SD (n 5 6).

836 LIN ET AL.


antibody against GP Ib antibody was detectedaround nanoparticles, suggesting that glycocalicinhad indeed been conjugated onto the nanoparticles.

Specific interaction between P-selectin-coatedsurfaces and GP Iba-conjugated nanoparticles underflow conditions was further analyzed using fluores-

cent intensity measurements (Fig. 6). Approximately50% GP Iba nanoparticles adhered to P-selectin-coated surfaces, while about 22% of control nanopar-ticles and GP Ib nanoparticles preincubated withanti-GP Iba antibodies adhered, respectively. Thecontrol nanoparticles adhered minimally to P-selec-tin and P-selectin/anti-P-selectin surfaces, indicatingthat nonselective binding under shear stress wasminimal. Using antibodies against GPIb with GPIb-conjugated nanoparticles also showed insignificantlevels of nanoparticle adhesion onto P-selectin-coated surfaces, indicating the importance of GPIbbinding onto P-selectin. In addition, surfaces coatedwith anti-P-selectin displayed no specific interactionswith any of the nanoparticles. Confocal microscopy

Figure 3. Confocal images of nanoparticle uptake in ECs.The image represents cells incubated with polystyrenenanoparticles. Plasma membranes were dyed using TexasRed1 and imaged using a RITC filter. Fluorescent nano-particles were imaged using a FITC filter. The images rep-resent an overlay of RITC and FITC filters and were takenat Ex(l) 488 nm and Em(l) 543 nm. [Color figure can beviewed in the online issue, which is available at www.interscience.wiley.com.]

Figure 4. Effects of shear stress on the cellular uptake of100-nm nanoparticles by HAECs. Values were obtained af-ter 30 min of flow with nanoparticle solutions and repre-sented as mean 6 SD (n 5 4). * and ** indicate the signifi-cant differences when compared with the static samples(p < 0.05 and p < 0.001, respectively). [Color figure can beviewed in the online issue, which is available at www.interscience.wiley.com.]

Figure 5. Cytoviva images of GPIb conjugation onto 100-nm polystyrene nanoparticles. The fluorescent nanoparticleswere imaged using a FITC (green) filter (A), and image (B) represents the image obtained using both the FITC and TexasRed (Red) filters. Magnification is at 310. [Color figure can be viewed in the online issue, which is available at www.nterscience.wiley.com.]



analysis also confirmed the observation using a fluo-rometer: the highest fluorescence was detected onP-selectin-coated slides with adherent GP Iba-conju-gated nanoparticles (data not shown). These resultsindicate that GP Iba-conjugated nanoparticles dis-play an enhanced adhesion onto P-selectin-coatedsurfaces under physiological flow conditions.

Figure 6. Nanoparticle adhesion onto P-selectin and anti-P-selectin-coated surfaces using different nanoparticle sam-ples. Values represent mean 6 SD (n 5 6). *Significant dif-ferences when compared with the control nanoparticle sam-ples (p < 0.05). [Color figure can be viewed in the onlineissue, which is available at www.interscience.wiley.com.]

Figure 7. Cellular Adhesion of GP Ib conjugated nano-particles and control nanoparticles by HAECs under vary-ing shear stress. Values were obtained after 30 minutes offlow with nanoparticle solutions and represent mean 6 SD(n53). *Significant differences when compared with thecontrol nanoparticle static samples (p< 0.001). **Significantdifferences when compared with the control nanoparticles(p < 0.001) at each flow rate. [Color figure can be viewedin the online issue, which is available at www.interscience.wiley.com.]

Figure 8. Confocal images of cellular uptake of polystyrene nanoparticles in activated ECs under shear stress. (A) Repre-sents control, (B) GP Iba-conjugated nanoparticles, and (C) anti-GP Iba-conjugated nanoparticles. Plasma membranes weredyed using Texas Red1 and imaged using a RITC filter. Fluorescent nanoparticles were imaged using a FITC filter. Thethird image (far right) represents an overlay of RITC and FITC filters. All images were taken at Ex(l) 488 nm and Em(l)543 nm. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

838 LIN ET AL.


Glass slides seeded with activated HAECs wererun on the flow chamber to investigate the effects ofshear stress on the cellular uptake of nonconjugatedand GPIb-conjugated nanoparticles by activated en-dothelial cells. Results showed significant cellularuptake when GP Iba nanoparticles were perfused,while control nanoparticles displayed insignificantcellular uptake (Fig. 7). Nanoparticles conjugatedwith GP Iba displayed much higher uptake inHAECs activated with histamine as detected withincells, with a very low concentration of nanoparticlespresent in extracellular spaces (Fig. 8). Samples wereobserved at the middle point of each cell using thestack imaging option, with slice thickness set at 1lm. In contrast, nonconjugated nanoparticles weremostly detected in extracellular spaces, with minimalwithin the cells (Fig. 8). Control nanoparticles hadthe lowest cellular uptake under exposure to shearstress with almost no nanoparticles seen within thecells, and most nanoparticles were localized on thesurface of the glass slides. Negative control samplesby blocking the GP Iba ligand with an antibodyresulted in minimal cellular uptake (results notshown), indicating the specific interaction betweenGP Iba and P-selectin. These observations confirmthat GP Iba adhesion with P-selectin expressed inactivated HAECs is necessary for cellular uptakeunder conditions of fluid shear stress.

DISCUSSION

The purpose of this study was to characterize thecellular uptake of nanoparticles by endothelial cellsand to investigate a new strategy to improve cellularuptake and targeting of nanoparticles in activated orinflamed HAECs. Particle uptake studies showedthat decrease in particle size increased cellularuptake. In addition, under static conditions, optimalnanoparticle dosage was found to be 200 lg/mLwhile optimal incubation time was found to be30 min for 100-nm polystyrene nanoparticles. Ourresults have shown that nanoparticles conjugatedwith the GP Iba significantly increased the adhesionof nanoparticles to P-selectin-coated surfaces. Cellu-lar uptake studies conducted under fluid shear stressalso displayed increased nanoparticle uptake andtargeting ability when HAECs were activated withhistamine to express P-selectin. Results from our en-dothelial cell uptake and nanoparticle-targeting char-acterization studies suggest that the GP Iba canincrease nanoparticle-targeting abilities and endothe-lial cell uptake under conditions of fluid shear stress.

The cellular uptake of polystyrene nanoparticles inHAECs was investigated to determine the effects ofparticle size, concentration, incubation time, and lev-

els of shear stress. Carboxylated polystyrene par-ticles were chosen as they are frequently used asmodel particles (for ‘‘a proof-of-concept’’) to investi-gate drug-targeting strategies because of their nar-row size distribution, uniform composition, and non-degradation properties.20,21 Cellular uptake of poly-styrene nanoparticles by HAECs decreased as thesize of nanoparticles increased (Fig. 3). Similar toHAECs, cellular uptake of polystyrene particles(ranging from 1 to 6 lm) by both leukocytes andmacrophages reduced with an increase in the parti-cle size.22 In contrast to this study, Tabata andIkada23 found that the maximal cellular uptake ofpolystyrene particles (ranging from 0.5 to 5 lm insize) in mouse peritoneal macrophages was reachedwhen the particle size was 2 lm. However, thesestudies use particles range at a larger size and differ-ent cell types when compared with our studies. Forcellular uptake of nanoparticles as the same range asour studies, previous studies have shown that nano-particles below 200 nm are internalized in Caco-216,17 and HUVEC cells.16,24–26 These studies alsofound that cells uptake or internalize nanoparticlesthrough receptor-mediated endocytosis process.16,24–26

Understanding the mechanism of cellular nanoparticleuptake plays a major role in intracellular trafficking,thus, greatly modulating the effectiveness of anyinternalized drug.18,27–30 Results from our studiesand previous studies again confirm the advantageof nanoparticles when compared with micropar-ticles for intracellular drug delivery, by showingthat HAECs prefer to uptake only small-sized par-ticles. Other advantages of nanoparticles include lit-tle or no local inflammation and less risk of arterialocclusion.31,32

Besides size dependence, the uptake of nanopar-ticles in HAECs was also dose- and incubation time-dependent. HAEC uptake of polystyrene nanopar-ticles was saturated at low concentrations (200 lg/mL). Other studies using similarly sized polystyrenenanoparticles (100 nm) in Caco-2 cells also displayedcellular uptake proportional to dosage, but reachedsaturation at higher particle concentrations (500 lg/mL).16,17 The difference in results between this studyand previous studies may be due to different celltypes. Nevertheless, our result, combined with previ-ous data, suggest that a drug can be delivered usingnanoparticles with particle concentration-based dos-ages. In fact, increasing nanoparticle concentration inthe infusate has been shown to significantly increasearterial uptake of drug (antiproliferative reagents)-loaded nanoparticles, leading to the increased levelsof antiproliferative reagents at the arterial wall usingacute dog models.33–35

In addition to dose dependence, the uptake ofnanoparticles by cells is also dependent on the incu-bation or exposure time. Previous studies using



Caco-2 cells displayed a similar trend in cellularuptake, but reached saturation after 2 h of incuba-tion.16,17 In contrast, our study found that the uptakein endothelial cells was saturated after 30 min ofincubation. These results might imply that particleuptake in endothelial cells is more active, requiringlower incubation times, and reaching the saturationlimit more rapidly than Caco-2 cells. Interestingly, invivo studies using dogs have found that the nanopar-ticle arterial uptake was twofold for repeated shortinfusions of nanoparticle suspension (15 s 3 4 s)than a single prolonged infusion (60 s).34 Theseresults indicate that optimized concentration andtime could be used to enhance the particle deliveryand reduce the side effects of the drug.

Besides the particle composition, particle size, andcell type, cellular uptake of nanoparticles can also beaffected by other factors such as surface propertiesof nanoparticles, high concentration of serum, andlevels of shear stress. For instance, studies haveshown that cells do not uptake polystyrene nanopar-ticles as efficiently as PLGA nanoparticles, possiblydue to increased surface hydrophobicity orcharges.17,24 In addition to particle surface proper-ties, media with high serum content may also resultin excessive cell exocytosis of nanoparticles due tothe energy-dependent nature of the process.18 Thelevel of shear stress is another factor that can affectparticle adhesion to and uptaken by endothelialcells. Similar to our observation, others have foundthat adhesion of nanoparticles on surfaces and/orendothelial cells is inversely correlated to the levelsof shear stress.36,37

To enhance recruitment of nanoparticles to endo-thelial cells under physiological flow conditions, pre-vious studies use ‘‘endothelial-targeting particles’’coated with humanized antibodies against E- and P-selectins36,37 and especially ‘‘leukocyte-inspired’’ par-ticles.4,7–12 Micro and nanoparticles conjugated withantibodies against P-selectin,36,37 the selectin ligand,sialyl Lewisx (sLex),4,8,10–12,38 and LFA-139 werefound to adhere to surfaces coated with P-selectin,E-selectin, and ICAM-1 and activated endothelialcells. Similar to leukocyte-inspired nanoparticles, ournovel endothelial-targeting nanoparticles that mimicthe adhesion of platelets on activated endothelialcells were able to adhere to P-selectin-coated surfa-ces under physiological flow conditions. The nano-particles may also adhere better with von Willebrandfactor, the ligand of GP Iba, deposited on injuredvessel wall and expressed P-selectin on activatedHAECs when compared with leukocyte-inspirednanoparticles due to the higher binding strength ofthe platelet ligands under high shear stress condi-tions.15,40–42 Of interests in the leukocyte-mimickingapproaches, Eniola and Hammer4 developed novelendothelial-targeting nanoparticles that mimic leuko-

cyte adhesion onto endothelial cells using multiple-receptor targeting: the selectin ligand sLex and anantibody against intercellular cell adhesion mole-cule-1 (ICAM-1). These couple receptors have beeninvolved in the initial dynamic interaction (sLex) andthe firm arrest of leukocytes on the endothelium(ICAM-1). Therefore, nanoparticles with these dou-ble ligand bindings would have a great selective ad-hesion onto the inflamed endothelium.

One final note for this study is its limitations.There is a concern regarding the GP Iba and its rolein mediating adhesion onto vWF surfaces. Previousdiscussions have defined the role of GP Iba in adhe-sion of platelets toward vWF in vitro and invivo.13,14,43–45 Our study did not include vWF; how-ever, its binding action with GP Iba can occur undermuch higher fluid shear stress than found in P-selec-tin-mediated adhesion. Other studies have success-fully characterized this vWF-mediated adhesionunder shear stress,13,14,43–45 while less work has beendone on the role of P-selectin. Another limitation ofthis study is that nanoparticles made of polystyreneare nondegradable and cannot be used as a realdrug delivery carrier. Finally, the relationshipbetween the bond forces upon the GP Iba-vWF and/or GP Iba-P-selectin interaction and shear stresswith the size of the particles has not been investi-gated in this study. Although our studies have theseshortcomings, using polystyrene nanoparticles asmodel nanoparticles to confirm a proof-of-concept isnecessary to help us further in our next step todesign drug carriers that can selectively target endo-thelium under physiological flow conditions. Futurework on this project will include studies involvingvWF-coated surfaces and our GP Iba-conjugated anddrug-loaded biodegradable nanoparticles.

The authors acknowledge Chad Larson and Dr. SophiaPassy from the Department of Biology, University of Texasat Arlington, for their assistance with the confocal micro-scope. They are grateful to Hao Xu and Maham Rahimi fortheir assistance provided in the Nanomedicine and TissueEngineering Laboratory at UTA.

References

1. Davda J, Labhasetwar V. Characterization of nanoparticleuptake by endothelial cells. Int J Pharm 2002;233:51–59.

2. Kuldo JM, Ogawara KI, Werner N, Asgeirsdaottir SA, KampsJA, Kok RJ, Molema G. Molecular pathways of endothelialcell activation for (targeted) pharmacological intervention ofchronic inflammatory diseases. Curr Vasc Pharmacol2005;3:11–39.

3. Lutters BC, Leeuwenburgh MA, Appeldoorn CC, MolenaarTJ, Van Berkel TJ, Biessen EA. Blocking endothelial adhesionmolecules: A potential therapeutic strategy to combat athero-genesis. Curr Opin Lipidol 2004;15:545–552.

840 LIN ET AL.


4. Eniola AO, Hammer DA. In vitro characterization of leuko-cyte mimetic for targeting therapeutics to the endotheliumusing two receptors. Biomaterials 2005;26:7136–7144.

5. Sakhalkar HS, Hanes J, Fu J, Benavides U, Malgor R, BorrusoCL, Kohn LD, Kurjiaka DT, Goetz DJ. Enhanced adhesion ofligand-conjugated biodegradable particles to colitic venules.FASEB J 2005;19:792–794.

6. Yao SN, Wilson JM, Nabel EG, Kurachi S, Hachiya HL, Kura-chi K. Expression of human factor IX in rat capillary endothe-lial cells: Toward somatic gene therapy for hemophilia B.Proc Natl Acad Sci USA 1991;88:8101–8105.

7. Sakhalkar HS, Dalal MK, Salem AK, Ansari R, Fu J, KianiMF, Kurjiaka DT, Hanes J, Shakesheff KM, Goetz DJ. Leuko-cyte-inspired biodegradable particles that selectively andavidly adhere to inflamed endothelium in vitro and in vivo.Proc Natl Acad Sci USA 2003;100:15895–15900.

8. Zou X, Shinde Patil VR, Dagia NM, Smith LA, Wargo MJ,Interliggi KA, Lloyd CM, Tees DF, Walcheck B, LawrenceMB, Goetz DJ. PSGL-1 derived from human neutrophils is ahigh-efficiency ligand for endothelium-expressed E-selectinunder flow. Am J Physiol Cell Physiol 2005;289:C415–C424.

9. Burch EE, Shinde Patil VR, Camphausen RT, Kiani MF, GoetzDJ. The N-terminal peptide of PSGL-1 can mediate adhesionto trauma-activated endothelium via P-selectin in vivo. Blood2002;100:531–538.

10. Eniola AO, Hammer DA. Characterization of biodegradabledrug delivery vehicles with the adhesive properties of leuko-cytes II: Effect of degradation on targeting activity. Biomateri-als 2005;26:661–670.

11. Eniola AO, Rodgers SD, Hammer DA. Characterization ofbiodegradable drug delivery vehicles with the adhesive prop-erties of leukocytes. Biomaterials 2002;23:2167–2177.

12. Eniola AO, Hammer DA. Artificial polymeric cells for tar-geted drug delivery. J Control Release 2003;87:15–22.

13. Kumar RA, Dong JF, Thaggard JA, Cruz MA, Lopez JA,McIntire LV. Kinetics of GPIbalpha-vWF-A1 tether bondunder flow: Effect of GPIbalpha mutations on the associationand dissociation rates. Biophys J 2003;85:4099–4109.

14. Dong J, Schade AJ, Romo GM, Andrews RK, Gao S, McIntireLV, Lopez JA. Novel gain-of-function mutations of plateletglycoprotein IBalpha by valine mutagenesis in the Cys209-Cys248 disulfide loop. Functional analysis under statis anddynamic conditions. J Biol Chem 2000;275:27663–27670.

15. Chen J, Lopez JA. Interactions of platelets with subendothe-lium and endothelium. Microcirculation 2005;12:235–246.

16. Desai MP, Labhasetwar V, Walter E, Levy RJ, Amidon GL.The mechanism of uptake of biodegradable microparticles inCaco-2 cells is size dependent. Pharm Res 1997;14:1568–1573.

17. Win KY, Feng SS. Effects of particle size and surface coat-ing on cellular uptake of polymeric nanoparticles for oraldelivery of anticancer drugs. Biomaterials 2005;26:2713–2722.

18. Panyam J, Labhasetwar V. Dynamics of endocytosis and exo-cytosis of poly(D,L-lactide-co-glycolide) nanoparticles in vas-cular smooth muscle cells. Pharm Res 2003;20:212–220.

19. Nguyen KT, Eskin SG, Patterson C, Runge MS, McIntire LV.Shear stress reduces protease activated receptor-1 expressionin human endothelial cells. Ann Biomed Eng 2001;29:145–152.

20. Florence AT, Hillery AM, Hussain N, Jani PU. Factors affect-ing the oral uptake and translocation of polystyrene nanopar-ticles: Histological and analytical evidence. J Drug Target1995;3:65–70.

21. Hussain N. Fluorometric method for the simultaneous quan-titation of differently-sized nanoparticles in rodent tissue. IntJ Pharm 2001;214:55–61.

22. Ayhan H, Tuncel A, Bor N, Piskin E. Phagocytosis of mono-size polystyrene-based microspheres having different size

and surface properties. J Biomater Sci Polym Ed 1995;7:329–342.

23. Tabata Y, Ikada Y. Effect of the size and surface charge ofpolymer microspheres on their phagocytosis by macrophage.Biomaterials 1988;9:356–362.

24. Foged C, Brodin B, Frokjaer S, Sundblad A. Particle size andsurface charge affect particle uptake by human dendritic cellsin an in vitro model. Int J Pharm 2005;298:315–322.

25. Qaddoumi MG, Ueda H, Yang J, Davda J, Labhasetwar V,Lee VH. The characteristics and mechanisms of uptake ofPLGA nanoparticles in rabbit conjunctival epithelial celllayers. Pharm Res 2004;21:641–648.

26. Wiewrodt R, Thomas AP, Cipelletti L, Christofidou-Solomi-dou M, Weitz DA, Feinstein SI, Schaffer D, Albelda SM,Koval M, Muzykantov VR. Size-dependent intracellularimmunotargeting of therapeutic cargoes into endothelial cells.Blood 2002;99:912–922.

27. Qaddoumi MG, Gukasyan HJ, Davda J, Labhasetwar V, KimKJ, Lee VH. Clathrin and caveolin-1 expression in primarypigmented rabbit conjunctival epithelial cells: Role in PLGAnanoparticle endocytosis. Mol Vis 2003;9:559–568.

28. Muro S, Wiewrodt R, Thomas A, Koniaris L, Albelda SM,Muzykantov VR, Koval M. A novel endocytic pathwayinduced by clustering endothelial ICAM-1 or PECAM-1.J Cell Sci 2003;116:1599–1609.

29. Muro S, Cui X, Gajewski C, Murciano JC, Muzykantov VR,Koval M. Slow intracellular trafficking of catalase nanopar-ticles targeted to ICAM-1 protects endothelial cells from oxi-dative stress. Am J Physiol Cell Physiol 2003;285:C1339–C1347.

30. Muro S, Koval M, Muzykantov V. Endothelial endocyticpathways: Gates for vascular drug delivery. Curr Vasc Phar-macol 2004;2:281–299.

31. Panyam J, Labhasetwar V. Biodegradable nanoparticles fordrug and gene delivery to cells and tissue. Adv Drug DelivRev 2003;55:329–347.

32. Vasir JK, Labhasetwar V. Polymeric nanoparticles for genedelivery. Exp Opin Drug Deliv 2006;3:325–344.

33. Labhasetwar V, Song C, Humphrey W, Shebuski R, Levy RJ.Arterial uptake of biodegradable nanoparticles: Effect of sur-face modifications. J Pharm Sci 1998;87:1229–1234.

34. Song C, Labhasetwar V, Cui X, Underwood T, Levy RJ. Arte-rial uptake of biodegradable nanoparticles for intravascularlocal drug delivery: Results with an acute dog model. J Con-trol Release 1998;54:201–211.

35. Guzman LA, Labhasetwar V, Song C, Jang Y, Lincoff AM,Levy R, Topol EJ. Local intraluminal infusion of biodegrad-able polymeric nanoparticles. A novel approach for pro-longed drug delivery after balloon angioplasty. Circulation1996;94:1441–1448.

36. Blackwell JE, Dagia NM, Dickerson JB, Berg EL, Goetz DJ.Ligand coated nanosphere adhesion to E- and P-selectinunder static and flow conditions. Ann Biomed Eng2001;29:523–533.

37. Dickerson JB, Blackwell JE, Ou JJ, Shinde Patil VR, Goetz DJ.Limited adhesion of biodegradable microspheres to E- and P-selectin under flow. Biotechnol Bioeng 2001;73:500–509.

38. Eniola AO, Willcox PJ, Hammer DA. Interplay between roll-ing and firm adhesion elucidated with a cell-free systemengineered with two distinct receptor-ligand pairs. Biophys J2003;85:2720–2731.

39. Eniola AO, Krasik EF, Smith LA, Song G, Hammer DA. I-do-main of lymphocyte function-associated antigen-1 mediatesrolling of polystyrene particles on ICAM-1 under flow. Bio-phys J 2005;89:3577–3588.

40. Nesbitt WS, Mangin P, Salem HH, Jackson SP. The impact ofblood rheology on the molecular and cellular events underly-ing arterial thrombosis. J Mol Med 2006;84:989–995.



41. Andrews RK, Shen Y, Gardiner EE, Berndt MC. Platelet ad-hesion receptors and (patho)physiological thrombus forma-tion. Histol Histopathol 2001;16:969–980.

42. Grabowski EF. Thrombolysis, flow, and vessel wall interac-tions. J Vasc Interv Radiol 1995;6(6, Part 2, Suppl):25S–29S.

43. Dong JF, Moake JL, Nolasco L, Bernardo A, Arceneaux W,Shrimpton CN, Schade AJ, McIntire LV, Fujikawa K, LopezJA. ADAMTS-13 rapidly cleaves newly secreted ultralargevon Willebrand factor multimers on the endothelial surfaceunder flowing conditions. Blood 2002;100:4033–4039.

44. Kumar RA, Moake JL, Nolasco L, Bergeron AL, Sun C,Dong JF, McIntire LV. Enhanced platelet adhesion andaggregation by endothelial cell-derived unusually large mul-timers of von Willebrand factor. Biorheology 2006;43:681–691.

45. Simon DI, Chen Z, Xu H, Li CQ, Dong J, McIntire LV, Ballan-tyne CM, Zhang L, Furman MI, Berndt MC, Lopez JA. Plate-let glycoprotein ibalpha is a counterreceptor for the leukocyteintegrin Mac-1 (CD11b/CD18). J Exp Med 2000;192:193–204.

842 LIN ET AL.


shear-regulated uptake of nanoparticles by endothelial cells and development of...

Documents