Specific Targeting of HER2 Overexpressing Breast Cancer Cells with Doxorubicin-Loaded Trastuzumab-Modified Human Serum Albumin Nanoparticles Marion G. Anhorn, †,§ Sylvia Wagner, ‡,§ Jo ¨rg Kreuter, † Klaus Langer, † and Hagen von Briesen* ,‡ Fraunhofer Institute for Biomedical Engineering, Department of Cell Biology and Applied Virology, 66386 St. Ingbert, Germany, and Institute of Pharmaceutical Technology, Biocenter of Johann Wolfgang Goethe-University, 60438 Frankfurt/Main, Germany. Received February 25, 2008; Revised Manuscript Received September 23, 2008 Specific targeting of tumor cells to achieve higher drug levels in tumor tissue and to overcome cardiotoxic and other secondary effects is the major goal in cancer therapy. With trastuzumab as a humanized monoclonal antibody binding, the HER2 receptor specific targeting is possible. In the present study, target-oriented nanoparticles based on biodegradable human serum albumin (HSA) loaded with cytostatic drug doxorubicin were developed. The surface of the nanoparticles was modified by covalent attachment of trastuzumab. HER2 overexpressing breast cancer cells showed a good cellular binding and uptake of these nanoparticles. The specific transport of the cytostatic drug doxorubicin with this nanoparticulate formulation into the HER2 overexpressing breast cancer cells, their release, and biological activity was demonstrated. The results indicate that these cell-type specific drug-loaded nanoparticles could achieve an improvement in cancer therapy. To our knowledge, this is the first study demonstrating a specific trastuzumab-based targeting of HER2 overexpressing breast cancer cells with doxorubicin- loaded nanoparticles. INTRODUCTION Cytostatic drugs are very useful substances in the fight of tumor diseases. However, the main disadvantages of conven- tional chemotherapy are modest tumor response and dose limiting side effects because of nonspecific action of drugs to all fast proliferating tissues (1). One possibility to avoid these disadvantages is the development of colloidal drug carrier systems (2). In this regard, nanoparticles show a high drug loading efficiency with minor drug leakage as well as the ability to circumvent multidrug resistance paired with good storage stability. Using human serum albumin (HSA) as particle forming polymer has the following advantages (3): human serum albumin is a physiological starting material and particles made of HSA are biodegradable. Additionally, the production of these particles is easy and reproducible (4). Functional groups are located on the surface of the prepared particles. Consequently, covalent linkage, for example, with drug targeting ligands is possible. Depending on the carrier system, a differentiation between passive and active drug targeting can be made (5). Malignant tumors induce an increased angiogenesis with the consequence of hypervascularization consisting of leaky vasculature in combination with poor lymphatic drainage (6). The pore size in the vasculature ranges from 200 to 600 nm in diameter. These characteristics lead to an enhanced extravasation and accumula- tion of carrier systems such as liposomes and nanoparticles. This phenomenon is called “enhanced permeability and retention (EPR) effect”. Consequently, long-circulating nanoparticles with poly(ethylene glycol) modifications on their surface are able to conduct passive tumor targeting (7). A further approach is the covalent linkage of drug targeting ligands such as monoclonal antibodies on the surface of the drug carrier systems to specifically target malignant cells (8-10). Thus, an active targeting with the advantage of an efficient accumulation of drugs in tumor tissue with higher drug levels in target cells can be achieved. Concerning drug targeting ligands, the focus of many groups is mainly on monoclonal antibodies (mAb) (11). Trastuzumab (Herceptin), a humanized monoclonal antibody targeting the HER2 receptor, was approved by the FDA in 1998. HER2 is a member of a family of four transmembrane epidermal growth factor receptor (EGFR) tyrosine kinases and is overexpressed in a variety of human cancer cells such as breast, lung, and ovarian carcinomas (12). HER2 overexpression is an adverse prognostic factor in women with breast cancer (13). Trastu- zumab administered as single agent leads to response rates ranging from 15% to 30% (13). Increased response rates can be achieved by using the antibody in combination with conventional chemotherapy. In this context, some clinical trials showed that an anthracycline-containing regimen leads to a greater benefit in women whose tumors overexpress HER2 than regimes without anthracyclines (14). The main drawback of the combination of trastuzumab and anthracyclines is that both trastuzumab and anthracyclines show cardiotoxic effects. That is the reason why this antibody-drug combination is only used in clinical trials where physicians have the possibility to monitor the heart function of the patients. On the other hand, it could be expected that trastuzumab-modified nanoparticles as drug carrier system for anthracyclines could reduce these side effects due to the ability of the particles to transport the embedded drug specifically to the tumor cells (15). Furthermore, the antibody concentration of the drug carrier system is far below the concentrations used in a therapeutic trastuzumab regime and a therapeutic antibody effect is not intended. The antibody only has a ligand function within such a nanoparticulate system. Upon arrival at the tumor tissue, trastuzumab molecules which are covalently attached to the particle surface by poly(ethylene * Corresponding author. Dr. Hagen von Briesen, Fraunhofer Institute for Biomedical Engineering, Ensheimer Strasse 48, D-66386 St. Ingbert, Germany. Phone: +49-(0)6894-980-286. Fax: +49-(0)6894-980-400. E-mail: [email protected]. † Institute of Pharmaceutical Technology. ‡ Fraunhofer Institute for Biomedical Engineering. § Both authors contributed equally to this work. Bioconjugate Chem. 2008, 19, 2321–2331 2321 10.1021/bc8002452 CCC: $40.75 2008 American Chemical Society Published on Web 10/21/2008
Transcript
Specific Targeting of HER2 Overexpressing Breast Cancer Cells withDoxorubicin-Loaded Trastuzumab-Modified Human Serum AlbuminNanoparticles
Marion G. Anhorn,†,§ Sylvia Wagner,‡,§ Jorg Kreuter,† Klaus Langer,† and Hagen von Briesen*,‡
Fraunhofer Institute for Biomedical Engineering, Department of Cell Biology and Applied Virology, 66386 St. Ingbert,Germany, and Institute of Pharmaceutical Technology, Biocenter of Johann Wolfgang Goethe-University, 60438 Frankfurt/Main,Germany. Received February 25, 2008; Revised Manuscript Received September 23, 2008
Specific targeting of tumor cells to achieve higher drug levels in tumor tissue and to overcome cardiotoxic andother secondary effects is the major goal in cancer therapy. With trastuzumab as a humanized monoclonal antibodybinding, the HER2 receptor specific targeting is possible. In the present study, target-oriented nanoparticles basedon biodegradable human serum albumin (HSA) loaded with cytostatic drug doxorubicin were developed. Thesurface of the nanoparticles was modified by covalent attachment of trastuzumab. HER2 overexpressing breastcancer cells showed a good cellular binding and uptake of these nanoparticles. The specific transport of the cytostaticdrug doxorubicin with this nanoparticulate formulation into the HER2 overexpressing breast cancer cells, theirrelease, and biological activity was demonstrated. The results indicate that these cell-type specific drug-loadednanoparticles could achieve an improvement in cancer therapy. To our knowledge, this is the first studydemonstrating a specific trastuzumab-based targeting of HER2 overexpressing breast cancer cells with doxorubicin-loaded nanoparticles.
INTRODUCTION
Cytostatic drugs are very useful substances in the fight oftumor diseases. However, the main disadvantages of conven-tional chemotherapy are modest tumor response and doselimiting side effects because of nonspecific action of drugs toall fast proliferating tissues (1). One possibility to avoid thesedisadvantages is the development of colloidal drug carriersystems (2). In this regard, nanoparticles show a high drugloading efficiency with minor drug leakage as well as the abilityto circumvent multidrug resistance paired with good storagestability. Using human serum albumin (HSA) as particle formingpolymer has the following advantages (3): human serum albuminis a physiological starting material and particles made of HSAare biodegradable. Additionally, the production of these particlesis easy and reproducible (4). Functional groups are located onthe surface of the prepared particles. Consequently, covalentlinkage, for example, with drug targeting ligands is possible.
Depending on the carrier system, a differentiation betweenpassive and active drug targeting can be made (5). Malignanttumors induce an increased angiogenesis with the consequenceof hypervascularization consisting of leaky vasculature incombination with poor lymphatic drainage (6). The pore sizein the vasculature ranges from 200 to 600 nm in diameter. Thesecharacteristics lead to an enhanced extravasation and accumula-tion of carrier systems such as liposomes and nanoparticles. Thisphenomenon is called “enhanced permeability and retention(EPR) effect”. Consequently, long-circulating nanoparticles withpoly(ethylene glycol) modifications on their surface are able toconduct passive tumor targeting (7). A further approach is the
covalent linkage of drug targeting ligands such as monoclonalantibodies on the surface of the drug carrier systems tospecifically target malignant cells (8-10). Thus, an activetargeting with the advantage of an efficient accumulation ofdrugs in tumor tissue with higher drug levels in target cells canbe achieved.
Concerning drug targeting ligands, the focus of many groupsis mainly on monoclonal antibodies (mAb) (11). Trastuzumab(Herceptin), a humanized monoclonal antibody targeting theHER2 receptor, was approved by the FDA in 1998. HER2 is amember of a family of four transmembrane epidermal growthfactor receptor (EGFR) tyrosine kinases and is overexpressedin a variety of human cancer cells such as breast, lung, andovarian carcinomas (12). HER2 overexpression is an adverseprognostic factor in women with breast cancer (13). Trastu-zumab administered as single agent leads to response ratesranging from 15% to 30% (13). Increased response rates canbe achieved by using the antibody in combination withconventional chemotherapy. In this context, some clinical trialsshowed that an anthracycline-containing regimen leads to agreater benefit in women whose tumors overexpress HER2 thanregimes without anthracyclines (14). The main drawback of thecombination of trastuzumab and anthracyclines is that bothtrastuzumab and anthracyclines show cardiotoxic effects. Thatis the reason why this antibody-drug combination is only usedin clinical trials where physicians have the possibility to monitorthe heart function of the patients. On the other hand, it couldbe expected that trastuzumab-modified nanoparticles as drugcarrier system for anthracyclines could reduce these side effectsdue to the ability of the particles to transport the embeddeddrug specifically to the tumor cells (15). Furthermore, theantibody concentration of the drug carrier system is far belowthe concentrations used in a therapeutic trastuzumab regime anda therapeutic antibody effect is not intended. The antibody onlyhas a ligand function within such a nanoparticulate system. Uponarrival at the tumor tissue, trastuzumab molecules which arecovalently attached to the particle surface by poly(ethylene
* Corresponding author. Dr. Hagen von Briesen, Fraunhofer Institutefor Biomedical Engineering, Ensheimer Strasse 48, D-66386 St. Ingbert,Germany. Phone: +49-(0)6894-980-286. Fax: +49-(0)6894-980-400.E-mail: [email protected].
† Institute of Pharmaceutical Technology.‡ Fraunhofer Institute for Biomedical Engineering.§ Both authors contributed equally to this work.
Bioconjugate Chem. 2008, 19, 2321–2331 2321
10.1021/bc8002452 CCC: $40.75 2008 American Chemical SocietyPublished on Web 10/21/2008
glycol) cross-linker chains will lead to the entry of the drug-loaded particles into the tumor cells (9, 16). Due to thebiodegradability of human serum albumin, the particles aredegraded within the cells. Thereupon, the released cytostaticdrug specifically kills the tumor cells without acting in nontargetcells such as heart tissue. Therefore, this approach could be ananswer to the crucial combination of trastuzumab and doxoru-bicin with the possibility to offer this effective therapy to womenwith HER2 overexpressing cancer cells.
EXPERIMENTAL PROCEDURES
Human serum albumin (HSA, fraction V, purity 96-99%),glutaraldehyde 8% aqueous solution, and human IgG antibodywere obtained from Sigma (Steinheim, Germany). Doxorubicinwas obtained from Sicor (Milan, Italy). 2-Iminothiolane (Traut’sreagent) and D-Salt Dextran desalting columns were purchasedfrom Pierce (Rockford, USA). Trastuzumab (Herceptin) wasobtained from Roche (Mannheim, Germany). The succinimidylester of methoxy poly(ethylene glycol) propionic acid with anaverage molecular weight of 5.0 kDa (mPEG5000-SPA) andthe cross-linker poly(ethylene glycol)-R-maleimide-ω-NHS esterwith an average molecular weight of 5.0 kDa (NHS-PEG5000-Mal) were purchased from Nektar (Huntsville, USA). Allreagents were of analytical grade and used as received.
Preparation of Doxorubicin-Loaded Nanoparticles. Anamount of 160 mg HSA was dissolved in 4 mL purified waterand the solution was filtered through a 0.22 µm cellulose acetatemembrane filter (Schleicher & Schuell, Dassel, Germany). Analiquot (500 µL) of this solution was added to 200 µL of a0.5% (w/v) aqueous stock solution of doxorubicin. To thismixture, 300 µL of purified water was added. For lower drugloading of the particles, the 0.5% (w/v) doxorubicin stocksolution was diluted to 0.05% (w/v) or 0.025% (w/v), respec-tively, before mixing with the aqueous HSA solution. In orderto adsorb doxorubicin to human serum albumin in solution, themixture was incubated under stirring (550 rpm) for 2 h at roomtemperature (17). For the preparation of nanoparticles bydesolvation, 3 mL ethanol (96%, v/v) was added continuously(1 mL/min) with a tubing pump (Ismatec IPN, Glattbrugg,Switzerland). After protein desolvation, an aliquot of 4.7 or11.75 µL 8% glutaraldehyde solution, respectively, was addedto induce particle cross-linking (corresponding to 40% or 100%stoichiometric protein cross-linking, respectively). The cross-linking was performed for 24 h under constant stirring at ambienttemperature. Aliquots (2.0 mL) of the resulting nanoparticleswere purified by two cycles of differential centrifugation (16 100g, 12 min) and redispersion. Within the first cycle, redispersionwas performed with 2.0 mL purified water, whereas in thesecond cycle, nanoparticles were redispersed with phosphatebuffer (pH 8.0) to a volume of 500 µL using a vortexer andultrasonication. The polymer content of the doxorubicin-loadednanoparticles was determined by gravimetry.
The collected supernatants were used to determine the non-entrapped doxorubicin by HPLC. The content of entrappeddoxorubicin was calculated from the difference between totaldoxorubicin and unbound drug. For the quantification ofdoxorubicin, a Merck Hitachi D7000 HPLC system equippedwith a LiChroCART 250-4 LiChrospher-100 RP-18 column(Merck, Darmstadt, Germany) was used. Separation wasobtained using a mobile phase of water and acetonitrile (70:30) containing 0.1% trifluoroacetic acid at a flow rate of 0.8mL/min. Doxorubicin was quantified by UV (250 nm) andfluorescence detection (excitation 560 nm, emission 650 nm).
Surface Modification of Drug-Loaded Nanoparticles. Drug-loaded HSA nanoparticles were prepared as described aboveand were modified as follows: One milliliter of HSA nanopar-ticle suspension dispersed in phosphate buffer (pH 8.0) was
incubated with 250 µL of mPEG5000-SPA solution (60 mg/mL in phosphate buffer pH 8.0) or poly(ethylene glycol)-R-maleimide-ω-NHS ester, respectively, for 1 h at 20 °C underconstant shaking (Eppendorf thermomixer, 600 rpm). Thenanoparticles were purified by centrifugation and redispersionas described above. The content of the nanoparticles wasdetermined by gravimetry.
For the thiolation step of the antibodies, trastuzumab and IgGwere dissolved in phosphate buffer pH 8.0 at a concentrationof 1.0 mg/mL. For the introduction of thiol groups, trastuzumabor IgG, respectively, were incubated with a 50-fold molar excessof 2-iminothiolane solution (c ) 1.14 mg/mL; 40.2 µL) for 2 has previously described by Steinhauser et al. (16). The antibodieswere purified by size exclusion chromatography (SEC, D-SaltDextran desalting column). The resulting solution containedthiolated antibody (trastuzumab or IgG, respectively) at aconcentration of about 500 µg/mL. For the coupling reaction,1.0 mL of the sulfhydryl-reactive nanoparticle suspension wasincubated with 1.0 mL of the thiolated trastuzumab or IgG,respectively, to achieve a covalent linkage between antibodyand the nanoparticle system. For the preparation of samples withadsorptively attached antibody, 1.0 mL of the mPEG5000-SPAmodified nanoparticles were incubated with 1.0 mL of thiolatedtrastuzumab or IgG, respectively. The incubation of all sampleswas performed for 12 h at 20 °C under constant shaking (600rpm). The samples were purified from unreacted antibody bycentrifugation and redispersion as described earlier. To deter-mine unbound antibody, the resulting supernatants were col-lected and analyzed by size exclusion chromatography (SEC)on a SWXL column (7.8 mm × 30 cm) in combination with aTSKgel SWXL guard column (6 mm × 4 cm) (Tosoh Bioscience,Stuttgart, Germany) using phosphate buffer (pH 6.6) as eluentat a flow rate of 1.0 mL/min. Aliquots of 20.0 µL were injected,and the eluent fraction was monitored by detection at 280 nm.In order to calibrate the SEC system for molecular weight,globular protein standards were used. The amount of antibodybound to the nanoparticle surface was calculated as differencebetween the amount of antibody obtained after thiolation andpurification and the amount of antibody determined in thesupernatant obtained after the conjugation step.
Nanoparticle Characterization. Nanoparticles were analyzedwith regard to particle diameter and polydispersity by photoncorrelation spectroscopy (PCS) using a Malvern Zetasizer3000HSA (Malvern Instruments Ltd., Malvern, UK). The zetapotential was measured with the same instrument by LaserDoppler microelectrophoresis. Prior to both measurements, thesamples were diluted with purified water. Particle content wasdetermined by gravimetry. For this purpose, 50.0 µL of thenanoparticle suspension was pipetted into an aluminum weighingdish and dried for 2 h at 80 °C. After 30 min of storage in anexsiccator, the samples were weighed on a microbalance(Sartorius, Germany).
Cell Culture. For all experiments, HER2 overexpressingbreast cancer cells SK-Br-3 (ATCC, LGC Promochem, Wesel)were used. The cells were cultured at 37 °C and 7.5% CO2 inMcCoy’s 5A medium (PromoCell, Heidelberg) supplementedwith 10% fetal calf serum (Invitrogen, Karlsruhe) and antibiotics(50 U/mL penicillin and 50 µg/mL streptomycin; Invitrogen,Karlsruhe).
As a control cell line, HER2 weakly expressing MCF7 cellswere used. The cells were cultured at 37 °C and 7.5% CO2 inRPMI 1640 medium (Invitrogen, Karlsruhe) supplemented with10% fetal calf serum (Invitrogen, Karlsruhe) and antibiotics (50U/mL penicillin and 50 µg/mL streptomycin; Invitrogen,Karlsruhe).
Cellular Binding of Nanoparticles. HER2 overexpressingSK-Br-3 cells were cultured in 24 well plates (Greiner,
Frickenhausen) and treated with the different nanoparticleformulations for 4 h at 37 °C. After this incubation, the cellswere washed twice with PBS (Invitrogen, Karlsruhe), thentrypsinized and harvested. After washing with PBS and fixingwith FACS-Fix (10 g/L PFA and 8.5 g/L NaCl in PBS, pH7.4), flow cytometry (FACS) analysis was performed with10 000 cells/sample, using FACSCalibur and CellQuest Prosoftware (Becton Dickinson, Heidelberg). Due to a greenautofluorescence of these nanoparticles at 488/520 nm, thisFACS analysis was possible.
For control experiment, HER2 weakly expressing MCF7 cellswere cultured in 24 well plates (Greiner, Frickenhausen) andtreated with the different nanoparticle formulations for 4 h at37 °C. The sample preparation for flow cytometry analysis wasperformed as described before.
Cellular Uptake and Intracellular Distribution. Cellularuptake and intracellular distribution of the nanoparticles werestudied by confocal laser scanning microscopy. SK-Br-3 cellswere cultured on glass slides and treated with the differentnanoparticle formulations for 4 h at 37 °C. After that incubationtime, the cells were washed twice with PBS and the cellmembranes were stained with 50 ng/µL Concanavalin AAlexaFluor 594 (Invitrogen, Karlsruhe) for 2 min. Cells werefixed with cold ethanol for 30 min. After fixation, the cells werewashed and then embedded in Vectashield HardSet MountingMedium (Axxora, Grunberg). The confocal microscopy studywas performed with an Axiovert 200 M microscope with a 510NLO Meta device (Zeiss, Jena), MaiTai femtosecond or anargon ion laser and the LSM Image Examiner software. Thegreen autofluorescence of the nanoparticles was also used here.
Cellular Doxorubicin Recovery. Doxorubicin recovery wasperformed with flow cytometry (FACS) analysis. SK-Br-3 cellswere cultured in 24 well plates (Greiner, Frickenhausen) andtreated with the different nanoparticle formulations or freedoxorubicin for 4 h at 37 °C. After this incubation, the cellswere washed and incubated with fresh medium for differentperiods of time, and afterward trypsinized and harvested. Flowcytometry (FACS) analysis was performed after fixation withFACS-Fix (10 g/L PFA and 8.5 g/L NaCl in PBS, pH 7.4) with10 000 cells/sample, using FACSCalibur and CellQuest Prosoftware (Becton Dickinson, Heidelberg). The red fluorescenceof doxorubicin at 488/590 nm was used.
Biological Activity. Cell viability was determined in SK-Br-3 cells using WST-1 assay (Cell Proliferation Reagent; RocheDiagnostics) based upon the absorption measurement of for-mazan formation. Briefly, cells were cultured in 6 well platesand incubated with the different nanoparticulate formulationsin PBS or PBS containing 100 µM chloroquine for 4 h at 37
°C. An amount of 10 000 cells/well were seeded in 96 wellplates and further incubated for 7 d at 37 °C. The determinationof cell viability was carried out after addition of WST-1 reagentand formazan measurement as described in the manufacturer’sinstruction manual.
For control experiment, HER2 weakly expressing MCF7 cellswere used. The cell viability assay was performed as describedbefore.
Statistical Analysis. Statistical analysis was made using thetwo-tailed Mann-Whitney U Test, which is equivalent to theWilcoxon rank sum test (18-20). Statistical significance wasdefined at P < 0.01 values.
RESULTS AND DISCUSSION
Particle Preparation and Characterization. In the first stepof the study, doxorubicin-loaded nanoparticles (Dox-NP) wereprepared by using a well-established desolvation method (17).The nanoparticles were stabilized by two different amounts ofglutaraldehyde. Therefore, nanoparticles were achieved with40% (Dox-NP 40%) or 100% (Dox-NP 100%) stochiometriccross-linking of the particle matrix. Using 2-iminothiolane, thiolgroups were introduced into the monoclonal antibody trastu-zumab and the control antibody IgG. Finally, the drug-loadedparticles were activated with a heterobifunctional poly(ethyleneglycol)-R-maleimide-ω-NHS ester (NHS-PEG5000-Mal) or amonofunctional succinimidyl ester of methoxy poly(ethyleneglycol) propionic acid (mPEG5000-SPA), respectively. In thefirst case, the heterobifunctional cross-linker leads to a covalentlinkage between antibody and nanoparticle (Dox-NP-mAb)(Figure 1). In the second case, only an adsorptive bindingbetween antibody and nanoparticle was expected because of thenonreactive methoxy group at the end of the poly(ethylene)glycol chain (Dox-NP + thiol mAb).
The results of the physicochemical characterization arepresented in Tables 1 and 2 for 40% or 100% cross-linkednanoparticles, respectively. The resulting particles were char-acterized by particle diameters of 391.3 ( 14.0 nm and 385.2( 30.8 nm for 40% or 100% cross-linked nanoparticles,respectively. The polydispersity ranged between 0.053 ( 0.036and 0.094 ( 0.057 indicating a monodisperse particle sizedistribution. Depending on the cross-linking degree, the doxo-rubicin loading was 51.39 ( 1.29 µg/mg for 40% cross-linkednanoparticles and 57.90 ( 0.44 µg/mg for 100% nanoparticles.Independent of whether the surface of the particles was modifiedor not, the particles showed a zeta potential between -40.1 (2.3 mV and -46.4 ( 3.4 mV.
Covalent linkage of trastuzumab to the particle surface wasachieved with 14.92 ( 1.34 µg antibody/mg nanoparticle (40%
Figure 1. Covalent linkage of thiolated antibody to cross-linker-activated drug-loaded nanoparticle (black square ) doxorubicin).
Specific Targeting with Doxorubicin-Loaded Nanoparticles Bioconjugate Chem., Vol. 19, No. 12, 2008 2323
cross-linked nanoparticles) and 16.09 ( 3.71 µg trastuzumab/mg nanoparticle for the 100% cross-linked particles. With thecontrol antibody IgG, similar results were obtained: Particleswith a cross-linking degree of 40% showed a surface modifica-tion of 13.80 ( 2.30 µg antibody/mg nanoparticle, whereas morehighly cross-linked particles resulted in a binding of 15.39 (3.60 µg IgG/mg nanoparticle on their surface.
In the case of preparations with antibodies bound byadsorption, a larger difference between the 40% and 100% cross-linked nanoparticles was observed: With a higher degree ofcross-linking, the amount of adsorptively attached antibodyincreased from 3.50 ( 1.13 µg/mg (40%) to 7.52 ( 5.29 µg/mg (100%) for trastuzumab and from 0.93 ( 1.14 µg/mg (40%)to 2.73 ( 1.26 µg/mg (100%) for IgG. This may be due to acertain amount of covalent antibody binding by residues ofmonovalent bound glutaraldehyde. Under all preparation condi-tions, in the case of trastuzumab a more pronounced tendencyto unspecific binding to particle surfaces was observed.
When comparing the results of nanoparticles with a covalentlinkage to samples with only an adsorptive antibody attachmentto the particle surface, it becomes obvious that the majority ofthe antibody molecules were covalently attached to the particlesurface by the heterobifunctional PEG spacer. Only a minorpart of the ligands were adsorbed to the nanoparticles or boundby monovalent glutaraldehyde.
Cellular Binding. HER2 overexpressing breast cancer cellsSK-Br-3 were incubated with the same concentration (0.1 mg/mL referred to nanoparticles concentration) of nanoparticles witha covalent linkage of trastuzumab (Dox-NP-trastuzumab) as wellas with samples with only an adsorptive attachment of trastu-zumab on the particle surface (Dox-NP + thiol trastuzumab).Both nanoparticulate formulations showed good cellular binding.There was no difference in binding efficiency between the Dox-NP-trastuzumab and the Dox-NP + thiol trastuzumab (data notshown). Under the chemical point of view, covalent bonding ismuch stronger than adsorptive bonding. Particularly, this is truewhen the nanoparticles are administered into the bloodstream.The adsorptively bound antibodies are subjected to desorptionprocesses. This phenomenon can be avoided by applying acovalent linkage between nanoparticle and antibody. Therefore,due to a supposed better stability of the covalent coupling, onlythe nanoparticle preparations with antibodies covalently linkedby heterobifunctional PEG spacers were used for further cellculture experiments.
The SK-Br-3 cells were incubated with Dox-NP-trastuzumabnanoparticles (nanoparticle concentration corresponding to 2 µg/mL trastuzumab modification) with either 40% or 100% cross-linking. Just as the modality of mAb coupling, the degree ofcross-linking had no influence on cellular binding efficiency(Figure 2a). Both nanoparticulate formulations, the 100% cross-linked as well as the 40% cross-linked Dox-NP-trastuzumab,displayed good cellular binding. Because there was no differencein cellular uptake between the 100% and 40% cross-linkednanoparticulate formulation (recognized on the same histogramshift, the same Y Mean fluorescence and the same percentageof positive cells, Figure 2a), further experiments were performedonly with 100% cross-linked nanoparticles.
Specific targeting of tumor cells to achieve higher drug levelsin tumor tissue and to overcome cardiotoxic and other secondaryeffects is the major goal in cancer therapy. With trastuzumabas a humanized monoclonal antibody, binding the HER2receptor specific targeting is possible. HER2 overexpressingbreast cancer cells SK-Br-3 were incubated with either nano-particles modified with specific trastuzumab on surface (nano-particle concentration corresponding to 2 µg/mL trastuzumabmodification) or nanoparticles with unspecific control mAb IgGon the surface (concentration referred to specific nanoparticleconcentration). As shown in Figure 2a, there was a specifictargeting with Dox-NP-trastuzumab (73.80% positive cells forthe 100% cross-linked nanoparticles and 73.07% positive cellsfor 40% cross-linked nanoparticles), whereas Dox-NP-IgGnanoparticles had a marginal cellular binding (5.57% positivecells). When the cells were incubated with increasing amountsof Dox-NP-trastuzumab, cellular binding increased in a con-centration-dependent manner (see histograms and table in Figure2b). Therefore, a highly specific targeting of HER2 overex-pressing cancer cells could be shown with this trastuzumab-modified Dox-NP.
As a control experiment, we did flow cytometry bindingstudies with the trastuzumab-modified nanoparticles on theHER2 weakly expressing MCF7 breast cancer cell line. TheMCF7 cells were incubated with either nanoparticles modifiedwith specific trastuzumab on surface or control nanoparticleswith IgG modification. As seen in Figure 3, trastuzumab-modified nanoparticles showed a marginal cellular binding, dueto the low expression of HER2. Therefore, these MCF7 cellsrepresented an optimal commonly used negative control. Thecomparison of the cellular binding of trastuzumab-modified
Table 1. Physicochemical Characteristics of Trastuzumab- and IgG-Modified Doxorubicin-Loaded HSA Nanoparticles with 40% Cross-LinkedParticles (n ) 3, mean ( SD)
Table 2. Physicochemical Characteristics of Trastuzumab- and IgG-Modified Doxorubicin-Loaded HSA Nanoparticles with 100% Cross-LinkedParticles (n ) 4, mean ( SD)
nanoparticles on HER2 overexpressing SK-Br-3 breast cancercells, showing a really specific cellular binding (Figure 2a), andHER2 weakly expressing MCF7 breast cancer cells, showing amarginal cellular binding (Figure 3), demonstrated the specificityof our nanoparticulate drug targeting system.
Cellular Uptake and Intracellular Distribution. The cel-lular uptake and intracellular distribution of these nanoparticulateformulations were shown by confocal laser scanning microscopy(CLSM). HER2 overexpressing breast cancer cells were incu-bated with nanoparticles with covalent coupled trastuzumab onparticle surface (Dox-NP-trastuzumab) as well as with controlnanoparticles (Dox-NP-mPEG) (Figure 4). Green fluorescenceof specifically uptaken Dox-NP-trastuzumab was visible withinthe inner part of the red-stained cell membrane (Figure 4f),whereas only some unspecific Dox-NP-mPEG stuck onto theouter part of the cell membrane (Figure 4c). Dox-NP-trastu-
zumab incubated cells were optically sliced in a stack of 0.75µm thickness each by confocal laser scanning microscopy toprove the intracellular uptake. The picture series is displayedas a gallery (Supporting Information supplement 1). An anima-tion of such a picture stack is shown in Supporting Informationsupplement 2 as well as the 3D projection of cells withinternalized Dox-NP-trastuzumab in Supporting Informationsupplement 3. These movies demonstrate the intracellular uptakeof the Dox-NP-trastuzumab.
In summary, this flow cytometry and CLSM data clarify thespecific targeting and intracellular uptake of Dox-NP-trastu-zumab.
Cellular Doxorubicin Recovery. In a next step, the specifictransport of the cytostatic drug doxorubicin with this nanopar-ticulate formulation into the cancer cells, their release, andbiological activity was proven. A first recovery experiment was
Figure 2. Specific cellular binding due to surface modification of nanoparticles with trastuzumab (40% or 100% cross-linked). (a) HER2 overexpressingSK-Br-3 cells were treated with Dox-NP-trastuzumab (40% or 100% cross-linked) or Dox-NP-IgG for 4 h at 37 °C. Flow cytometry (FACS)analysis was performed to quantify their cellular binding. The data are shown as histograms of the FL1-H-channel (autofluorescence of thenanoparticles) as well as a table with Y Mean fluorescence data and percentage of positive cells. Dark green: Dox-NP-trastuzumab (40% cross-linked). Light green: Dox-NP-trastuzumab (100% cross-linked). Red: Dox-NP-IgG. Blue: untreated control. (b) SK-Br-3 cells were incubated withdifferent Dox-NP-trastuzumab concentrations for 4 h at 37 °C. Flow cytometry (FACS) analysis was performed to quantify their cellular binding.The data are shown as histograms of the FL1-H-channel (autofluorescence of the nanoparticles) as well as a table with Y Mean fluorescence dataand percentage of positive cells. Green: 20 µg/mL referred to mAb concentration. Red: 10 µg/mL. Orange: 2 µg/mL. Yellow: 1 µg/mL. Brown: 0.2µg/mL. Blue: untreated control.
Figure 3. Control of cellular binding on HER2 weakly expressing MCF7 cells. HER2 weakly expressing MCF7 cells were treated with Dox-NP-trastuzumab or Dox-NP-IgG for 4 h at 37 °C. Flow cytometry (FACS) analysis was performed to quantify their cellular binding. The data areshown as histograms of the FL1-H-channel (autofluorescence of the nanoparticles) as well as a table with Y Mean fluorescence data. Green:Dox-NP-trastuzumab. Red: Dox-NP-IgG. Blue: untreated control.
Specific Targeting with Doxorubicin-Loaded Nanoparticles Bioconjugate Chem., Vol. 19, No. 12, 2008 2325
done as a flow cytometry analysis with the standard cell lineSK-Br-3 and different concentrations of free doxorubicin todemonstrate the possibility of distinguishing different intracel-
lular doxorubicin concentrations by flow cytometry. Use wasmade of the red fluorescence of doxorubicin in the FL2-Hchannel of the flow cytometer. Therefore, SK-Br-3 cells were
Figure 4. Cellular uptake and intracellular distribution of the nanoparticles studied by confocal laser scanning microscopy. SK-Br-3 cells werecultured on glass slides and treated with the different nanoparticle formulations for 4 h at 37 °C. The green autofluorescence of the nanoparticleswas used, and the cell membranes were stained with Concanavalin A AlexaFluor 594 (red). Pictures were taken within inner sections of the cells.(a-c) Incubation of the cells with control nanoparticles without mAb modification. (d-f) Incubation of the cells with the specific Dox-NP-trastuzumab.(a) and (d) display the green nanoparticles channel; (b) and (e) display the red cell membrane; (c) and (f) overlay of the both fluorescence channels.
Figure 5. Doxorubicin recovery studied by flow cytometry: possibility of distinguishing different intracellular doxorubicin concentrations and thedetection of nanoparticular transported doxorubicin. (a) HER2 overexpressing SK-Br-3 cells were treated with different concentrations of freedoxorubicin for 4 h at 37 °C. Flow cytometry (FACS) analysis was performed to quantify their cellular binding. The data are shown as histogramsof the FL2-H-channel (fluorescence of the doxorubicin) as well as a table with Y Mean fluorescence data and percentage of positive cells. Green:50 µM doxorubicin concentration. Red: 15 µM. Orange: 10 µM. Yellow: 5 µM. Brown: 1 µM. Blue: untreated control. (One representative experimentout of two independent experiments is shown.) (b) HER2 overexpressing SK-Br-3 cells were treated with different nanoparticulate formulations orfree doxorubicin for 4 h at 37 °C. Flow cytometry (FACS) analysis was performed to quantify their cellular binding. The data are shown ashistograms of the FL2-H-channel (fluorescence of the doxorubicin) as well as a table with Y Mean fluorescence data and percentage of positivecells. Orange: free doxorubicin. Green: Dox-NP-trastuzumab. Red: Dox-NP-IgG. Blue: untreated control. (One representative experiment out oftwelve independent experiments is shown.)
incubated with different doxorubicin concentrations for 4 h at37 °C. The results of flow cytometry analysis are shown inFigure 5a. A clear distinction between the different doxorubicinconcentrations is possible with flow cytometry and CellQuestPro software.
When the cells were incubated with doxorubicin-loadednanoparticles with specific trastuzumab on particle surface (NPconcentration corresponding to 15 µM doxorubicin concentra-tion) or with doxorubicin-loaded nanoparticles with unspecificIgG (NP concentration corresponding to 15 µM doxorubicin)or with the same concentration of free doxorubicin (15 µM),different histograms were detected on the FL2-H channel (Figure5b). Free doxorubicin showed the highest level of fluorescenceintensity, followed by Dox-NP-trastuzumab and then Dox-NP-IgG (see table of Y Mean fluorescence Figure 5b). This effectdemonstrated that different amounts of doxorubicin weretransported to the cells by the different nanoparticulate formula-tions or free doxorubicin. Due to its specific cellular uptake,Dox-NP-trastuzumab transported more doxorubicin to the cellsthan the unspecific Dox-NP-IgG. The better uptake of freedoxorubicin is a known phenomenon, but this is an unspecifictransport to all kinds of cells responsible for all adversesecondary effects during a doxorubicin-based cancer therapy.
However, it was possible to influence the transported doxo-rubicin concentration by nanoparticles different in doxorubicinloading. HER2 overexpressing SK-Br-3 cells were treated withthe specific Dox-NP-trastuzumab nanoparticles with an intra-particular doxorubicin concentration of 5.96 µg or 2.78 µgdoxorubicin/mg nanoparticle or with the unspecific control Dox-NP-IgG nanoparticles with the same intraparticular doxorubicinconcentration for 4 h at 37 °C. The flow cytometry data showeda better nanoparticulate cellular binding in the FL1-H-channel(Figure 6a) for the specific nanoparticulate formulations (over
90% positive cells) than the unspecific nanoparticulate formula-tions (approximately 1.3% positive cells), as already known.However, there was no difference in binding among the Dox-NP-trastuzumab nanoparticulate formulations with the differentintraparticular doxorubicin concentrations, as well as amongDox-NP-IgG, meaning that the intraparticular doxorubicinconcentration does not influence the cellular binding. However,there was a better transport of doxorubicin to the cells by higherloaded Dox-NP-trastuzumab than by lower loaded (see histo-gram and table in Figure 6b). This effect was also seen forunspecific nanoparticulate formulations. Thus, there was apossibility to influence the doxorubicin concentration transportedto the cell. This could also be done by a higher amount ofincubated nanoparticles, but a certain relation between thenumber of cells and the quantity of incubated nanoparticlesshould be maintained. Indeed, it is important for a nanopar-ticulate cancer therapy to administer highly loaded formulationsin order to achieve a small drug infusion volume.
In a further recovery experiment, HER2 overexpressing SK-Br-3 cells were treated with Dox-NP-trastuzumab (NP concen-tration corresponding to 7 µM doxorubicin concentration), Dox-NP-IgG (0.1 mg/mL NP concentration corresponding to 7 µMdoxorubicin concentration), or free doxorubicin (7 µM) for 4 hat 37 °C and were analyzed directly or further incubated withfresh medium for 20 or 44 h. Flow cytometry (FACS) analysiswas performed to quantify their cellular binding and toinvestigate the doxorubicin processing. The data are shown ashistograms of the FL2-H-channel for 4 h nanoparticle incubationonly (Figure 7a), 4 h nanoparticle incubation and 20 h incubationtime with medium (Figure 7b), and 4 h nanoparticle incubationand 44 h incubation time with medium (Figure 7c). For a betterunderstanding of the results, we drew two lines in Figure 7a-c.The first goes through the histogram maximum of the histogram
Figure 6. Doxorubicin recovery studied by flow cytometry. HER2 overexpressing SK-Br-3 cells were treated with different nanoparticulate formulationswith different intraparticular doxorubicin concentrations for 4 h at 37 °C. Flow cytometry (FACS) analysis was performed to quantify their cellularbinding. (a) The data are shown as histograms of the FL1-H-channel (autofluorescence of the nanoparticles) as well as a table with Y Meanfluorescence data and percentage of positive cells. (b) The data are shown as histograms of the FL2-H-channel (fluorescence of the doxorubicin)as well as a table with Y Mean fluorescence data and percentage of positive cells. Deep green: Dox-NP-trastuzumab nanoparticles with an intraparticulardoxorubicin concentration of 5.96 µg doxorubicin/mg nanoparticle. Light green: Dox-NP-trastuzumab with an intraparticular doxorubicin concentrationof 2.78 µg doxorubicin/mg nanoparticle. Red: control Dox-NP-IgG with an intraparticular doxorubicin concentration of 5.96 µg doxorubicin/mgnanoparticle. Orange: Dox-NP-IgG with an intraparticular doxorubicin concentration of 2.78 µg doxorubicin/mg nanoparticle.
Specific Targeting with Doxorubicin-Loaded Nanoparticles Bioconjugate Chem., Vol. 19, No. 12, 2008 2327
of free doxorubicin incubation analyzed after 4 h, and the secondgoes through the histogram maximum of the histogram of Dox-NP-trastuzumab after 4 h incubation. Both lines were elongatedand cross the histograms after further 20 h and 44 h incubation.First of all, the already known histogram formation of doxo-rubicin transport to the cells was recognized: free doxorubicinwas the best, followed by Dox-NP-trastuzumab and Dox-NP-IgG. However, if the cells were incubated with fresh mediumand analyzed after 20 or 44 h a faster regression of thedoxorubicin signal of free doxorubicin was recognized than ofDox-NP-trastuzumab. For the regression of the doxorubicinsignal of free doxorubicin, we can read off values fromapproximately 65 (4 h incubation) to 13 (further 44 h incuba-tion), which is a regression by a factor of 5. For the regressionof the doxorubicin signal of Dox-NP-trastuzumab, we can readoff values from approximately 13 (4 h incubation) to 8 (further44 h incubation), which is a regression by a factor of 1.6. Areason for this observation could be the known drug effluxpumps of the cells. Free doxorubicin is transported out of the
cells by these pumps, but nanoparticulate wrapped doxorubicinis saved. This sustained release effect could help to minimizethe needed doxorubicin quantity for the cancer therapy in thefuture and to reduce secondary effects.
Biological Activity. In addition to the cellular uptake of thenanoparticulate doxorubicin, the therapeutic effect of theseformulations was determined. HER2 overexpressing SK-Br-3cells were incubated with the specific Dox-NP-trastuzumab orthe unspecific Dox-NP-IgG in PBS for 4 h at 37 °C at aconcentration of 15 µM referred to doxorubicin concentration,and a further incubation time of 7 d followed. Cell viabilitywas determined using WST-1 assay based on the absorbancemeasurement of formazan formation. The formazan formationis only possible in the mitochondria of healthy cells; therefore,untreated cells set the 100% standard. After the incubation ofDox-NP-trastuzumab, the cells showed only a viability of 20.1%compared to cells incubated with Dox-NP-IgG reaching aviability of 56.5% (Figure 8a). Therefore, the Dox-NP-trastu-zumab offered a better specific cellular uptake as well as a bettercytotoxic activity.
To test whether the nanoparticles are entrapped within theendosomes or lysosomes and could not release their loadedcytostatic drug, chloroquine was used. Chloroquine is a weakbasic reagent, which rapidly diffuses into the cells and ac-cumulates in acidic compartments. This accumulation leads toan inhibition of lysosomal enzymes because of the pH increase.Lysosomal enzymes only work at acidic pH. Furthermore,chloroquine inhibits the fusion of endosome and lysosome, soentrapped agents are no longer degraded. The accumulation ofchloroquine in endosomal vesicles induces an osmotic swellingof them, and therefore, the endosomes are destabilized andrelease the entrapped agents. However, there was no significantdifference between the viability of the cells incubated withnanoparticulate formulations in either PBS (Figure 8a) or PBScontaining chloroquine (Figure 8b). After incubation with Dox-NP-trastuzumab, there were 20.1% of healthy cells in the caseof PBS incubations and 19.6% in the case of chloroquine use.The findings after incubation with Dox-NP-IgG were similar.There are 56.5% of healthy cells after incubation in PBS and51.7% after incubation in PBS containing chloroquine. Theseresults indicate that the nanoparticulate formulations are notentrapped in the endosomes or lysosomes, and the release ofthe cytostatic drug is not prohibited respectively.
In order to exclude whether or not these biological effectsresult from the trastuzumab surface modification of the nano-particles and are mainly induced by the biological effect oftrastuzumab itself, we compared doxorubicin-loaded trastu-zumab-modified nanoparticles and unloaded trastuzumab-modi-fied nanoparticles. As shown in Figure 9, the doxorubicin-loadedtrastuzumab-modified nanoparticles had, as already known, acytotoxic effect on cells, but the unloaded trastuzumab-modifiednanoparticles did not alter the cell viability. Therefore, we canconclude that the trastuzumab modification of the particles hadno influence on cell viability and the doxorubicin componentof the particulate system must have caused the biological activityof our nanoparticulate system.
To demonstrate the specificity of our nanoparticulate system,we incubated the HER2 weakly expressing control cell lineMCF7 with either the specific trastuzumab-modified nanopar-ticles or the unspecific IgG-modified nanoparticles (NP con-centration corresponding to 1 µM doxorubicin). Both nanopar-ticulate formulations showed the same cytotoxicity due to anunspecific nanoparticle uptake of both nanoparticle formulations(Figure 10). These results verify the specificity and the biologicalactivity of our newly designed ligand targeted nanoparticulateformulation based on human serum albumin with doxorubicinloading.
Figure 7. Doxorubicin recovery by flow cytometry: intracellulardoxorubicin processing. HER2 overexpressing SK-Br-3 cells weretreated with different nanoparticulate formulations or free doxorubicinfor 4 h at 37 °C and incubated with fresh medium for different periodsof time at 37 °C. Flow cytometry (FACS) analysis was performed toquantify their cellular binding. The data are shown as histograms ofthe FL2-H-channel (fluorescence of the doxorubicin) as well as a tablewith Y Mean fluorescence data. Orange: free doxorubicin. Green: Dox-NP-trastuzumab. Red: Dox-NP-IgG. Blue: untreated control. (a) 4 hnanoparticle incubation only. (b) 4 h nanoparticle incubation and 20 hincubation time with medium. (c) Four h nanoparticle incubation and44 h incubation time with medium. (One representative experimentout of four independent experiments is shown.)
In the future, nanoparticle development for cancer chemo-therapy will be a significant and expanding part of nanobio-technology. The search for new tumor targets, novel ligands,new strategies for targeting, and particle stabilization will occupyscience in the future to improve drug delivery at the tumor sitewhile decreasing toxicity to normal tissues.
There already exist a lot of nanostructured formulations inorder to reach this goal (for review, see Haley et al. (21)). Forexample, there are FDA-approved liposomal doxorubicin en-capsulations (Doxil/Caelyx and Myocet) where the anthracyclinepharmacokinetics are changed and cardiac risk is decreased butnot eliminated (22-25). Anthracycline toxicity to normal tissueis reduced by liposomal encapsulation. After extravasationthrough tumor endothelium, liposomes disintegrate and deliverdoxorubicin. Drug concentration has been measured 10 timeshigher in tumor tissue treated with Doxil compared to conven-tional free drug administration (26).
A further example is the first HSA-based nanoparticleformulation, Abraxane, approved by the FDA in 2005 and bythe EMEA in 2008. The 130 nm large nanoparticles containthe cytostatic drug paclitaxel. Due to the bad solubility ofpaclitaxel in water, the conventional drug preparation (Taxol)contains polyethylated castor oil (Cremophor EL) and ethanolas vehicles. The drug incorporation in nanoparticles follows a
new concept to improve drug solubility, with a variety ofadvantages compared to the standard paclitaxel therapy. There-fore, a nanoparticulate-bound paclitaxel infusion leads to a 33%increase in intratumoral concentrations and a 50% higher doseof delivered paclitaxel in comparison to a conventional paclitaxelinfusion. Additionally, since nanoparticulate-bound paclitaxelis solvent-free, the infusion time is 30 min in contrast to the3 h infusion for conventional taxol, and no premedication isrequired (27, 28).
In earlier studies, we have optimized and established adesolvation process to prepare human serum albumin nanopar-ticles (4). Previous studies have demonstrated that nanoparticlesbased on HSA were rapidly degraded after cellular uptake andled to a significant cellular accumulation of the entrapped drug(29). Therefore, we advanced this process to load human serumalbumin based nanoparticles with the antineoplastic agentdoxorubicin (17).
As already discussed, the specific targeting of cancer cellswill be a breakthrough in cancer therapy. The treatment withmonoclonal antibodies recognizing tumor-associated markersis a common strategy in cancer therapy. One of the mostpromising targets is the human epidermal growth factor recep-tor-2 (HER2), whose overexpression is strongly associated withpoor prognosis. HER2 is a transmembrane tyrosine kinase
Figure 8. Cell viability assay: comparison between Dox-NP-trastuzumab and Dox-NP-IgG. HER2 overexpressing SK-Br-3 cells were incubatedwith the Dox-NP-trastuzumab or Dox-NP-IgG at the same concentration in (a) PBS or (b) PBS containing chloroquine for 4 h at 37 °C. Afterwashing, the cells were further incubated for 7 d at 37 °C. The cell viability was determined after addition of WST-1 reagent and measurement offormazan formation, as described in the manufacturer’s instructions manual. Untreated cells set the 100% standard. (internal control of each experimentn ) 9, one representative experiment out of three independent experiments is shown for experiment (a))* *: The two samples are significantlydifferent (P < 0.01, two-tailed Mann-Whitney U Test equivalent to the Wilcoxon rank sum test).
Figure 9. Cell viability assay: comparison between Dox-NP-trastu-zumab and NP-trastuzumab. HER2 overexpressing SK-Br-3 cells wereincubated with the Dox-NP-trastuzumab or NP-trastuzumab at the sameconcentration for 4 h at 37 °C. After washing, the cells were furtherincubated for 7 d at 37 °C. The cell viability was determined afteraddition of WST-1 reagent and measurement of formazan formation,as described in the manufacturer’s instructions manual. Untreated cellsset the 100% standard. (internal control of each experiment n ) 9)* *:The two samples are significantly different (P < 0.01, two-tailedMann-Whitney U Test equivalent to the Wilcoxon rank sum test).
Figure 10. Cell viability assay: comparison between Dox-NP-trastu-zumab and Dox-NP-IgG. HER2 weakly expressing MCF7 cells wereincubated with the Dox-NP-trastuzumab or Dox-NP-IgG at the sameconcentration for 4 h at 37 °C. After washing, the cells were furtherincubated for 7 d at 37 °C. The cell viability was determined afteraddition of WST-1 reagent and measurement of formazan formation,as described in the manufacturer’s instructions manual. Untreated cellsset the 100% standard. (Internal control of each experiment n ) 9)* *:The two samples are significantly different (P < 0.01, two-tailedMann-Whitney U Test equivalent to the Wilcoxon rank sum test).
Specific Targeting with Doxorubicin-Loaded Nanoparticles Bioconjugate Chem., Vol. 19, No. 12, 2008 2329
growth factor receptor, which is only marginally expressed inadult tissue (30), but is overexpressed in approximately 30%of human gastric, lung, and breast carcinoma. Trastuzumab(Herceptin), a humanized monoclonal antibody directed againstthe extracellular domain of HER2, is the only HER2-targetedtherapy approved by the FDA for treatment of breast cancer(31). However, the best results in therapy were achieved whentrastuzumab was given in combination with other cytotoxicagents. Because of its enhanced expression on tumor cells, itsextracellular accessibility and its ability to internalize afterantibody binding, HER2 represents a suitable target for tumortherapy with cell-specific nanoparticles. In a former study, wedemonstrated an improvement of the specific uptake into HER2overexpressing cells with trastuzumab surface-modified humanserum albumin based nanoparticles (9, 16).
In the present study, we combined the specific trastuzumab-based nanoparticles with the cytostatic drug loading efficiency,to achieve a specific drug delivery system. Thus, we canconclude that we have developed a novel specific drug carriersystem, which allowed us a specific transport of doxorubicinto HER2 overexpressing breast cancer cells. The duration ofthe uptaken doxorubicin within these cells was elongated andthe drug was better saved against efflux, a known effect ofnanoparticulate incorporated cytostatic drugs. Therefore, aminimization of needed doxorubicin volume in cancer therapyis presumed. Concerning the specificity and the minimized druginfusion volume, a reduction of the secondary side effects canbe estimated. Due to the mAb directed specificity of thenanoparticles, our novel drug carrier system should be moreefficient compared to the already known PEGylated liposomaldoxorubicin carrier (Doxil or Caelyx). Additionally, nanopar-ticles show a higher loading capacity with a lower tendency ofdrug leakage than liposomes. Furthermore, a better stability ofnanoparticles compared to liposomal formulations is known.
These trastuzumab-modified nanoparticles efficiently bind toand are specifically internalized in HER2 overexpressing breastcancer cells. They release the incorporated drug effectively. Thecombination of specific targeting with drug loading in thesenanoparticulate formulations should lead to an improvement incancer therapy.
ACKNOWLEDGMENT
The authors thank Daniel Sauer for his skilful technicalassistance. This work was financially supported by the GermanBundesministerium fur Bildung und Forschung (BMBF) (Project13N8671).
Supporting Information Available: Additional graphics andanimated files as described in the text. Please refer to the pdffor descriptions. This material is available free of charge viathe Internet at http://pubs.acs.org.
LITERATURE CITED
(1) Cirstoiu-Hapca, A., Bossy-Nobs, L., Buchegger, F., Gurny, R.,and Delie, F. (2007) Differential tumor cell targeting of anti-HER2 (Herceptin) and anti-CD20 (Mabthera) coupled nanopar-ticles. Int. J. Pharm. 331, 190–196.
(2) Couvreur, P., and Vauthier, C. (2006) Nanotechnology: intel-ligent design to treat complex disease. Pharm. Res. 23, 1417–1450.
(3) Weber, C., Coester, C., Kreuter, J., and Langer, K. (2000)Desolvation process and surface characterisation of proteinnanoparticles. Int. J. Pharm. 194, 91–102.
(4) Langer, K., Balthasar, S., Vogel, V., Dinauer, N., von Briesen,H., and Schubert, D. (2003) Optimization of the preparationprocess for human serum albumin (HSA) nanoparticles. Int.J. Pharm. 257, 169–180.
(5) Yokoyama, M., and Okano, T. (1996) Targetable drug carriers:present status and a future perspective. AdV. Drug DeliVery ReV.21, 77–80.
(6) Maeda, H., Wu, J., Sawa, T., Matsumura, Y., and Hori, K.(2000) Tumor vascular permeability and the EPR effect inmacromolecular therapeutics: a review. J. Controlled Release65, 271–284.
(7) Greenwald, R. B. (2001) PEG drugs: an overview. J. ControlledRelease 74, 159–171.
(8) Nobs, L., Buchegger, F., Gurny, R., and Allemann, E. (2004)Current methods for attaching targeting ligands to liposomes andnanoparticles. J. Pharm. Sci. 93, 1980–1992.
(9) Wartlick, H., Michaelis, K., Balthasar, S., Strebhardt, K.,Kreuter, J., and Langer, K. (2004) Highly specific HER2-mediated cellular uptake of antibody-modified nanoparticles intumour cells. J. Drug Targeting 12, 461–471.
(10) Dinauer, N., Balthasar, S., Weber, C., Kreuter, J., Langer, K.,and von Briesen, H. (2005) Selective targeting of antibody-conjugated nanoparticles to leukemic cells and primary T-lymphocytes. Biomaterials 26, 5898–5906.
(11) Adams, G. P., and Weiner, L. M. (2005) Monoclonal antibodytherapy of cancer. Nat. Biotechnol. 23, 1147–1157.
(12) Harries, M., and Smith, I. (2002) The development and clinicaluse of trastuzumab (Herceptin). Endocr. Relat. Cancer 9, 75–85.
(13) Valabrega, G., Montemurro, F., and Aglietta, M. (2007)Trastuzumab: mechanism of action, resistance and future per-spectives in HER2-overexpressing breast cancer. Ann. Oncol.18, 977–984.
(14) Ferretti, G., Felici, A., Papaldo, P., Fabi, A., and Cognetti,F. (2007) HER2/neu role in breast cancer: from a prognosticfoe to a predictive friend. Curr. Opin. Obstet. Gynecol. 19, 56–62.
(15) Pereverzeva, E., Treschalin, I., Bodyagin, D., Maksimenko,O., Langer, K., Dreis, S., Asmussen, B., Kreuter, J., andGelperina, S. (2007) Influence of the formulation on thetolerance profile of nanoparticle-bound doxorubicin in healthyrats: focus on cardio- and testicular toxicity. Int. J. Pharm.337, 346–356.
(16) Steinhauser, I., Spankuch, B., Strebhardt, K., and Langer, K.(2006) Trastuzumab-modified nanoparticles: Optimisation ofpreparation and uptake in cancer cells. Biomaterials 27, 4975–4983.
(17) Dreis, S., Rothweiler, F., Michaelis, M., Cinatl, J., Jr., Kreuter,J., and Langer, K. (2007) Preparation, characterisation andmaintenance of drug efficacy of doxorubicin-loaded human serumalbumin (HSA) nanoparticles. Int. J. Pharm. 341, 207–214.
(18) Siegel, S., Castellan, N. J. (1988) Nonparametric statistics forthe behavioral sciences (2nd edition), McGraw-Hill Education,New York.
(19) Sachs, L. (1992) Angewandte Statistik (7.Auflage), Springer-Verlag, Berlin.
(20) Heath, D. (1995) An Introduction to Experimental Design andStatistics for Biology, UCL Press, London.
(21) Haley, B., and Frenkel, E. (2008) Nanoparticles for drugdelivery in cancer treatment. Urol. Oncol. 26, 57–64.
(22) Working, P. K., Newman, M. S., Sullivan, T., and Yarrington,J. (1999) Reduction of the cardiotoxicity of doxorubicin in rabbitsand dogs by encapsulation in long-circulating, pegylated lipo-somes. J. Pharmacol. Exp. Ther. 289, 1128–1133.
(23) Waterhouse, D. N., Tardi, P. G., Mayer, L. D., and Bally,M. B. (2001) A comparison of liposomal formulations ofdoxorubicin with drug administered in free form: changingtoxicity profiles. Drug Saf. 24, 903–920.
(24) Gabizon, A., Shmeeda, H., and Barenholz, Y. (2003)Pharmacokinetics of pegylated liposomal Doxorubicin: reviewof animal and human studies. Clin. Pharmacokinet. 42, 419–436.
(25) O’Brien, M. E., Wigler, N., Inbar, M., Rosso, R., Grischke,E., Santoro, A., Catane, R., Kieback, D. G., Tomczak, P.,Ackland, S. P., Orlandi, F., Mellars, L., Alland, L., and
Tendler, C. (2004) Reduced cardiotoxicity and comparableefficacy in a phase III trial of pegylated liposomal doxorubicinHCl (CAELYX/Doxil) versus conventional doxorubicin forfirst-line treatment of metastatic breast cancer. Ann. Oncol.15, 440–449.
(26) Lasic, D., Martin, F. (1995) Stealth Liposomes, in Handbooksin Pharmacology and Toxicology (Lasic, D., and Martin, F., Eds.)pp 22-41, CRC Press, Boca Raton, FL.
(27) Gradishar, W. J., Tjulandin, S., Davidson, N., Shaw, H., Desai,N., Bhar, P., Hawkins, M., and O’Shaughnessy, J. (2005) PhaseIII trial of nanoparticle albumin-bound paclitaxel compared withpolyethylated castor oil-based paclitaxel in women with breastcancer. J. Clin. Oncol. 23, 7794–7803.
(28) Desai, N., Trieu, V., Yao, Z., Louie, L., Ci, S., Yang, A.,Tao, C., De, T., Beals, B., Dykes, D., Noker, P., Yao, R., Labao,E., Hawkins, M., and Soon-Shiong, P. (2006) Increased antitumor
activity, intratumor paclitaxel concentrations, and endothelial celltransport of cremophor-free, albumin-bound paclitaxel, ABI-007,compared with cremophor-based paclitaxel. Clin. Cancer Res.12, 1317–1324.
(29) Wartlick, H., Spankuch-Schmitt, B., Strebhardt, K., Kreuter,J., and Langer, K. (2004) Tumour cell delivery of antisenseoligonuclceotides by human serum albumin nanoparticles. J.Controlled Release 96, 483–495.
(30) Press, M. F., Cordon-Cardo, C., and Slamon, D. J. (1990)Expression of the HER-2/neu proto-oncogene in normal humanadult and fetal tissues. Oncogene 5, 953–962.
(31) Tokunaga, E., Oki, E., Nishida, K., Koga, T., Egashira, A.,Morita, M., Kakeji, Y., and Maehara, Y. (2006) Trastuzumaband breast cancer: developments and current status. Int. J. Clin.Oncol. 11, 199–208.
BC8002452
Specific Targeting with Doxorubicin-Loaded Nanoparticles Bioconjugate Chem., Vol. 19, No. 12, 2008 2331
