Colloidal DNA Carriers for Direct Localization in Cell Compartments

School of Chemical and Biomedical Engineering, Nanyang Technological University, Singapore 637457, Singapore, School of Materials Science and Engineer...
0 downloads 4 Views 4MB Size
Download PDF
Biomacromolecules 2010, 11, 1779–1784

1779

Colloidal DNA Carriers for Direct Localization in Cell Compartments by pH Sensoring Uta Reibetanz,*,†,‡ Min Hui Averil Chen,‡ Shaillender Mutukumaraswamy,‡ Zi Yen Liaw,‡ Bernice Hui Lin Oh,‡ Subbu Venkatraman,§ Edwin Donath,| and Bjo¨rn Neu*,‡ School of Chemical and Biomedical Engineering, Nanyang Technological University, Singapore 637457, Singapore, School of Materials Science and Engineering, Nanyang Technological University Singapore 639798, Singapore, and Institute of Medical Physics and Biophysics, Medical Faculty, University of Leipzig, 04107 Leipzig, Germany Received March 3, 2010; Revised Manuscript Received May 11, 2010

Multifunctional colloidal microparticles allow the integration of various active agents as well as reporter molecules into one system without interfering combining delivery and sensing functions. In this study, calcium carbonate particles were functionalized with fluorescein isothiocyanate-labeled poly(allylamine hydrochloride) (FITC-PAH) allowing particle localization in cell compartments of different pH. Plasmid DNA (pEGFP-C1 and pDsRed1-N1) as a reporter agent for drug release in the cytoplasm and rhodamine-B-isothiocyanate-labeled protamine (RITCPRM) were integrated into biocompatible and biodegradable PRM/DXS multilayers. The uptake and processing of the particles by HEK293T/17 cells were investigated via flow cytometry and confocal laser scanning microscopy. The presented data show a clear correlation between the fluorescence intensity of the FITC-labeled core, that is, the particle localization after cellular uptake, and the expression of fluorescent proteins by the cells without further cell staining. In conclusion, this particle design allows the simultaneous study of particle location and processing to monitor the transport and release of active agents and should thus be an invaluable tool for the study and design of nano- and microcarrier systems.

Introduction Microparticles and capsules, layer-by-layer (LbL)-coated with biocompatible and biodegradable polyelectrolytes, have received much attention as drug delivery systems in medicine and pharmacy.1 The main advantage of such multilayer-coated carriers is the integration of various agents in one system without interfering. Most recent efforts have been directed toward the application of these particles and capsules as carrier systems, either for active agents or for biosensors for cellular applications.1-5 Microparticles and capsules fabricated by the LbL technique allow the incorporation of a wide range of active and sensing agents either into the multilayer or into the core.6-8 Sensing molecules have been used mostly for tracking particles and capsules inside and outside of cells.9,10 Additional surface modifications such as antibody-functionalized lipid layers or viruses have also been assembled to control specific cell interaction and subsequent cellular uptake.11-13 Even though quite a number of studies have been carried out regarding the fabrication and (bio)functionalization of LbLcoated particles and capsules,14-16 so far only little is known on how these particles are processed in cells. The combination of transport and sensing functions in a single particle should thus help to gain a more detailed understanding of transport * To whom correspondence should be addressed. Tel.: +49-341-9715703 (U.R.); +65-6790-6951 (B.N.). Fax: +49-341-9715709 (U.R.); +65-67911761 (B.N.). E-mail: [email protected] (U.R.). † Present address: Institute of Medical Physics and Biophysics, Medical Faculty, University of Leipzig, Ha¨rtelstraβe 16-18, D-04107 Leipzig, Germany, Tel. +49-341-9715737, Fax: +49-341-9715709. ‡ School of Chemical and Biomedical Engineering, Nanyang Technological University. § School of Materials Science and Engineering, Nanyang Technological University. | University of Leipzig.

and releasing processes of particles and capsules within the cell. Such knowledge is crucial for the understanding and, thus, manipulation of the multilayer degradation for a controlled, timedependent release of active agents into specific cell compartments. Previously, LbL particles were functionalized with a pHdependent fluorophore, fluorescein isothiocyanate (FITC), for particle localization in endolysosomes or cytoplasm17 or pEGFPC1 and pDsRed1-N1 were assembled into the multilayer as a reporter for successful delivery in the cytoplasm.18 In these studies, flow cytometry was employed to detect and quantify particles in different cell compartments, which are distinguishable by their different pH. However, due to the low expression rate of the reporter agent caused by a low particle release into the cytoplasm, this method does not allow relating the release rate of the particles into the cytoplasm to the release of the active agents. In this study, a combination of two fluorescence methods were established for particle tracking: (1) the quantitative investigation of particle/cell interaction by flow cytometry and (2) the qualitative confocal imaging of expressing and nonexpressing cells. Supplementary to the particle localization by flow cytometry, confocal microscopy allows a direct investigation of the expressing or nonexpressing behavior of the cells in relation to the precise location of the particles in the cell compartments. Thus, our particle design provides a new tool for monitoring the particles throughout their processing within the cell, with the aim to ensure a more successful delivery of active agents.

Materials and Methods Materials. Poly(ally amine) hydrochloride (PAH, MW ) 70 kDa), protamine sulfate salt from herring (PRM, MW)4 kDa), rhodamine isothiocyanate (RITC), fluorescein isothiocyanate (FITC), phosphatebuffered saline (PBS), dimethyl sulfoxide (DMSO), sodium hydrogen

10.1021/bm100237r  2010 American Chemical Society Published on Web 06/15/2010

1780

Biomacromolecules, Vol. 11, No. 7, 2010

Reibetanz et al.

Table 1. PRM/DXS and Plasmid DNA Coating Schemes (a) CaCO3 particles as plasmid DNA transporter (b) Quantification of particles in cells using flow cytometry (c) Correlation of FITC intensity and protein expression via CLSM

carbonate (Na2CO3), calcium chloride (CaCl2), citric acid, disodium hydrogen phosphate, paraformaldehyde (PFA), and FITC-conjugated cholera-toxin B were obtained from Sigma-Aldrich (Singapore). Dextran sulfate (DXS, MW ) 70 kDa) was purchased from MP Biomedicals (Singapore). Sephadex G25 chromatography gel and trypan blue (TB) were obtained from Sigma-Aldrich (Singapore) and plasmid DNA pEGFP-C1 and pDsRed1-N1 were purchased from Clontech. HEK 293T/17 cells were obtained from ATCC, and Dulbecco’s modified Eagle medium (DMEM) and fetal calf serum (FCS) were purchased from PAA. Polyelectrolyte Labeling. The covalent binding of rhodamine isothiocyanate to protamine (RITC-PRM) was initiated in 0.1 M carbonate buffer (pH 8.5) by dropwise addition of a RITC-methanol solution. After incubating for 2 d under constant stirring, RITC was removed by gel chromatography over sephadex 25G. The binding of fluorescein isothiocyanate to poly(allylamine hydrochloride) (FITCPAH) was conducted in 0.1 M sodium hydrogen carbonate buffer (pH 8.5) by adding FITC dissolved in DMSO. After 2 days incubation, the FITC-PAH was purified using a sephadex 25G column. The concentration of bound FITC was determined by UV/vis spectrometry taking an extinction coefficient of ε490 ) 80900 M-1 cm-1. One RITC molecule per one PRM molecule and one FITC molecule per 753 PAH monomers were labeled, respectively. The labeled polyelectrolytes were lyophilized and stored at 4 °C. Particle Preparation. For the preparation of CaCO3 cores, 0.33 M NaCO3, and 0.33 M CaCl2 solutions were used.19 FITC-labeled PAH (1 µL, 60 µg/µL) was added to 250 µL of NaCO3 solution and mixed with 250 µL of CaCO3 solution. This mixture was incubated for 5-10 min until the precipitation process was finished and washed four times in double distilled water. A concentration of 60 µg PAH-FITC in 500 µL of NaCO3 and CaCl2 solution resulted in particle sizes of about 5-7 µm. Particle Coating. The particles were alternately incubated in PRM and DXS solution (4 mg/mL, 0.1 M NaCl) for 10 min under constant shaking. After each coating step the particles were centrifuged and washed three times in 0.1 M NaCl. Plasmid DNA pEGFP-C1 and pDsRed1-N1 were coated in a 1 M NaCl solution, incubated for 30 min, and washed three times in 0.1 M NaCl. Particle/Cell Interaction. Particle/cell experiments were performed in 8-well chamber slides containing 5 × 105 HEK293T/17 cells seeded 1 day before particle incubation. A particle/cell ratio of 5:1 was used for uptake experiments. After incubating the particles for 2 days, they were investigated by means of flow cytometry and confocal laser scanning microscopy. Flow Cytometry (FCM). Fluorescence intensity of polyelectrolytecoated particles and green fluorescent cells was investigated by flow cytometry (FACSCalibur, Becton Dickinson, U.S.A.) with a laser excitation wavelength of 488 nm (FITC detection in FL1, RITC detection in FL2). Particles (104) or cells with particle interaction were detected in each measurement and were analyzed by WinMDI2.9 software. After coincubating cells with particles for two days, they were trypsinized and washed with PBS and immediately measured via flow cytometry. The measurement of the pH-dependent particle FITC fluorescence intensity was performed in citrate-phosphate buffer (CIP, 0.1 M citric acid solution, 0.2 M disodium hydrogen phosphate solution), with the pH adjusted between pH 2 and pH 9. Particles, stored separately from cell experiments in 0.1 M NaCl solution, were added to 300 mL of buffer solution and measured after a short incubation time of 1 min to prevent core dissolution under acidic conditions. For quenching experiments, particles as well as cell/particle samples were incubated in Trypan Blue (TB) for 30 min.

CaCO3| CaCO3| CaCO3| CaCO3|

[PRM/DXS]3/PRM/pEGFP/PRM [PRM/DXS]3/PRM/pDsRed/PRM [PRM/DXS]3/PRM/pDsRed/PRM PRM/DXS/[PRM-R/DXS]2/PRM/pDsRed/PRM

Confocal Laser Scanning Microscopy (CLSM). The CLSM images were obtained with a Zeiss LSM 510 Meta Laser Scanning Microscope equipped with a Plan-Apochromat 63x/1.4 oil objective. For the FITC and GFP fluorescence, the Ar/Kr laser was used, and for the RITC and DsRed fluorescence, the He/Ne laser was used. All images were obtained in sequential mode between frames to separate green and red emission. The cells were fixed with PFA and stained with FITC-labeled cholera toxin B to visualize particles within the cells. Particles suspended in CIP buffer (pH 2-pH 9) served as a control for the pHdependent fluorescence intensity. After a short incubation time, the fluorescence intensity of 3-4 particles in selected cells, in supernatant, and in CIP buffer were analyzed by ImageJ and Origin6.1. For this purpose, a circular region of interest (ROI) was chosen and the ROI intensity distribution was analyzed.

Results and Discussion CaCO3 Particles as Plasmid DNA Transporter. The system employed in this study combines the transport, sensor, and release capabilities of FITC-PAH-labeled CaCO3 particles coated with (RITC-labeled) PRM and plasmid DNA. As previously shown,18 particles with a PRM/DXS multilayer coated on silica particles can be used as a plasmid DNA carrier. In comparison to silica particles, CaCO3 cores are biocompatible, can be functionalized during preparation, and can be more easily dissolved after multilayer coating to produce capsules.16 Initial efforts were directed toward testing the suitability of these CaCO3 particles as a microcarrier system. For this purpose, plasmid DNA pEGFP-C1 and pDsRed-N1 were incorporated in the biodegradable PRM/DXS multilayer as reporters (Table 1a), and the successful transport and release of the plasmid DNA was tested by the expression of GFP or DsRed. Figure 1a,b shows the expression of the fluorescent proteins clearly, proving the successful cellular uptake, release of the particles into the cytoplasm, subsequent multilayer degradation, and release of the intact plasmid DNA molecules. Quantification of Particles in Cells Using Flow Cytometry. To study the cell/particle interaction, pDsRed1N1 was incorporated into a PRM/DXS multilayer on FITC cores (Table 1b). Figure 2 shows the FITC intensity distribution of cells incubated with and without particles. For cells incubated with particles, the intensity profile can be separated into two distinct regions. Region 1 corresponds to cells that do not show a significant increase in the fluorescence intensity and thus indicate no interaction with particles, whereas cells in region 2 demonstrate a significant increase in the fluorescence intensity, indicating interaction with the fluorescently labeled particles. Two forms of interaction are possible: particles can be either attached to the cell surface or incorporated into the cell. About 12% of the cells show an increase in the fluorescence intensity and, thus, particle interaction. Taking into consideration that we employed a particle to cell ratio of 5:1, the amount of cells showing particle interaction is in the same range as previously reported for silica particles.18 Region 2 can be further divided into two regions, 2-1 and 2-2 (Figure 3a), representing different FITC fluorescence intensities. In comparison, the fluorescence intensity distribution

DNA Microcarriers for Direct Localization in Cells

Biomacromolecules, Vol. 11, No. 7, 2010

1781

Figure 1. CLSM images of HEK293T/17 cells after uptake of PRM/DXS-coated CaCO3 particles functionalized with either (a) pDsRed1-N1 or (b) pEGFP-C1.

Figure 2. Fluorescence intensities of HEK293T/17 cells after 2 days incubation with CaCO3 particles with PRM/DXS and pDsRed1-N1 as multilayer constituents measured via FCM. About 12% of the cells are interacting with particles (region 2).

of the particles in the supernatant, that is, without cell interaction, is shown in Figure 3c (black line, before quenching). The fluorescence intensity distribution of these particles corresponds to region 2-2 in Figure 3a, suggesting that the cells interact mostly with no more than one particle. Otherwise, the intensity distribution in Figure 3a should demonstrate a broader peak with higher fluorescence intensities. Figure 3d shows the intensity of the FITC functionalized particles as a function of the pH. Comparing this curve with the fluorescence intensity peaks in Figure 3a allows estimating the pH values of the respective particle environments. In region 2-1 this results in a pH of 4.8 ( 0.2 and in region 2-2 in a pH of 6.9 ( 0.2. Whereas these observations clearly indicate that region 2-1 should correspond to the pH of the endolysosome, region 2-2 requires further analysis. One problem of the FITC pH sensorbased detection of particle processing in cells by flow cytometry is the inability to distinguish between particle locations exposed to the same pH, that is, region 2-2 originates from particles either attached to the outer cell membrane or localized within the cytoplasm. To separate the two fractions, the FITC label of the core was used as localization marker in a fluorescencequenching assay. Trypan blue (TB) as a quenching agent can only affect those particles attached to the outer membrane, since it cannot cross the membrane of living cells.20 Thus, the cell sample was

incubated with TB to identify the surface-bound particles. The gray curves in Figure 3b (cell/particles interaction) and c (particles in supernatant) represent the intensity distributions after TB incubation. In Figure 3b, the right peak 2-2 is partially shifted to lower intensities while the left peak 2-1 appears unaffected by intensity changes. For the particles freely suspended in the supernatant, shown in Figure 3c, the intensity maximum after quenching is also shifted to a lower value with respect to untreated particles (black curve). The unchanged cell/ particle fraction in region 2-2 represents the cells with particles in the cell cytoplasm, whereas the shifted cell/particle fraction corresponds to particles attached to the outer membrane. The intensity distributions of regions 2-1 and 2-2 were than fitted to three Gaussian functions in Figure 3a and to four in Figure 3b, according to the expected number of peaks. The maxima were fixed at defined values. The maximum M1 of region 2-1 is given by the geometric mean value of region 2-1 at 45 ( 2 au, the maximum M2 of region 2-2 (unquenched and quenched intensity) is given by the particle geometric mean intensity (Figure 3c) at 225 ( 30 au (unquenched intensity) and shifts to the value of M1 at 45 ( 7 after TB quenching. M3 represents cells incorporating double particles with intensity at 500 ( 30 au. In Figure 3b, three Gaussian functions are not sufficient to describe the intensity distribution. A fourth Gaussian function with a maximum M4 has to be introduced in Figure 3b after TB

1782

Biomacromolecules, Vol. 11, No. 7, 2010

Reibetanz et al.

Figure 4. Dependence of the fluorescence intensity of FITC-PAHlabeled CaCO3 particles on the pH measured by means of CLSM. The fluorescence intensity at lower pH is characterized by a narrow peak at low intensities, whereas an increasing pH value results in a broader peak and a maximum shifted toward higher intensities.

Figure 3. Fluorescence intensities in region 2 (see Figure 2) measured via FCM (a) before and (b) after TB quenching; fluorescence intensities of the particles in the supernatant (c) before and after TB incubation and (d) the pH dependence of the mean fluorescence intensities.

quenching. The maximum of this function cannot be fixed. It is found between the intensity maxima of particles exposed to pH 4.8 and 6.9. It can be assumed that the limited intensity shift after TB quenching is caused by a partial restricted accessibility of the core fluorophores to particles strongly attached to the membrane. Thus, it is determined that 20 ( 3% of the particles are located in endolysosomes (M1, Figure 3a), 42 ( 4% in the cytoplasm (M2

and M3, Figure 3b), and 38 ( 3% are attached to the outer membrane (M4 and M1(b)-M1(a), Figure 3b). Direct Correlation of FITC Intensity and Protein Expression Using CLSM. The investigation of particle/cell interaction via flow cytometry allows quantifying the amount of particles localized within cell compartments of different pH. However, one problem with this method is the inability to correlate the particle location with the efficiency of the release of the active agent, represented by cells expressing the red or green fluorescent protein, due to the low number of expressing cells. Thus, CLSM analysis was also employed in another set of experiments to investigate if the FITC intensity in cells can be correlated to protein expression. As a reporter, pDsRed1-N1 and RITC-PRM were incorporated in multiple PRM/DXS layers on FITC-PAH-labeled CaCO3 cores (Table 1c). The particles were than incubated in CIP buffer for 1 min and images were recorded via CLSM. The results are shown in Figure 4. The particle intensity distribution is depicted depending on the pH values of the surrounding buffer in a range of pH 2-9. The intensity distribution changes from a narrow peak with a maximum at low intensities (pH 2-4) to a broad appearance with a maximum shifted to higher intensities (pH 7-9). In Figures 5 and 6, functionalized particles were incubated with cells and the fluorescence intensity was then quantified via CLSM. Figures 5a and 6a were recorded with the same amplification used for recording the CLSM images of the particles suspended in supernatant (data not shown) to calculate the pH-dependent intensity differences, whereas in Figures 5b and 6b the settings were adjusted to show the cell staining clearly. Each image in Figures 5 and 6 is divided into three panels, I-III. I and II represent the green and red fluorescence, respectively, and the overlay of both is shown in panel III. Figure 5 shows a cell after particle uptake, without any detectable expression of the red fluorescent protein. The CTx staining of the cell membrane confirms the internalization of the particle. In comparison, the images in Figure 6 represent cells expressing DsRed after particle uptake. Comparing the FITC intensities of the particles in Figures 5a-I and 6a-I clearly indicate that particles outside the cells and within DsRedexpressing cells have a significantly higher fluorescence intensity as compared to those incorporated in nonexpressing cells (Figure 6b-1, panel I). The mean values of the intensities of 3-4 particles in expressing and nonexpressing cells are shown in

DNA Microcarriers for Direct Localization in Cells

Biomacromolecules, Vol. 11, No. 7, 2010

1783

Figure 5. CLSM images of nonexpressing HEK293T/17 cells after particle interaction. The CaCO3 particles are coated with PRM-R/DXS and pDsRed1-N1. In the upper row (a), the images were recorded with the same settings as used for particles in the supernatant, and in the lower row (b), the same ROI was recorded at a higher amplification. The panels I-III represent the FITC and RITC fluorescence distribution as well as the overlay, respectively.

Figure 6. CLSM images of DsRed expressing HEK293T/17 cells after particle interaction. The CaCO3 particles are coated with PRM-R/DXS and pDsRed1-N1. In the upper row (a), the images were recorded with the same settings as used for particles in the supernatant, and in the lower row (b), the same ROI was recorded at a higher amplification. Panels I-III represent the FITC and RITC fluorescence distribution as well as the overlay, respectively.

the histograms in (Figure 7). Figure 7a depicts the intensity distribution of particles outside the cells as a control and the histograms in Figure 7b,c show the particle intensity distributions in DsRed-expressing cells and nonexpressing cells, respectively. The partial pixel increase at lower intensities (Figure 7a,b) is mainly caused by background intensity. Comparing the fluorescence intensities of Figure 7a,b with those in Figure 7d suggests that free particles as well as particles in expressing cells are exposed to a pH of about 8 ( 0.5, indicating that the internalized particles are released into the cytoplasm, followed by degradation of the biopolymer multilayer and release of the plasmid DNA. On the other hand, comparing the intensities in Figure 7c with those in 7d implies that particles,

which were taken up by nonexpressing cells are exposed to an acidic environment with a pH of 5 ( 0.5. This pH corresponds to the endolysosomes and is, thus, in agreement with the assumption that these particles remain in the endolysosomes where they are exposed to degrading enzymes that prevent the transport and release of the reporter agent, the plasmid DNA, into the cytoplasm for further processing.

Conclusion The multifunctionality of colloidal transport systems based on µm-sized particles or capsules allows combining transport and sensor functions in a single carrier: various active or reporting agents can be integrated into the multilayer and the core without

1784

Biomacromolecules, Vol. 11, No. 7, 2010

Reibetanz et al.

The fluorescent marker was used in two ways. Measuring the fluorescence intensity of each particle by flow cytometry allowed us to quantify the particles in the endolysosome and in the cytoplasm. In addition, confocal imaging was used to correlate the particle location with the activity of the reporter DNA. For both methods, the pH-dependent fluorescence intensity of the particles showed a difference of about factor 7 for the regions of interest, pH 7-8 and pH 4-5, allowing to estimate the pH of the particle environment. In conclusion, the combination of both methods provides additional information about the processing of the particles within cells. Only particles that were released into the cytoplasm were able to release the active agent successfully, whereas plasmid DNA in particles that were retained in endolysosomes was degraded, showing no activity. In summary, we demonstrated a new method to evaluate the colloidal transport systems in order to enhance their efficiency as potential drug carriers. The integrated pH sensor function and reporter markers in the multilayer help to better understand cell-carrier interaction, which is valuable in future studies of optimizing the processing of the carriers for transport and release of active agents within cells. Acknowledgment. This work was supported by grants from the Ministry of Education (Singapore) and the National Research Foundation (Singapore).

References and Notes

Figure 7. CLSM intensity histogram of particles in different environments: (a) particles located in the supernatant and (b) particles in DsRed expressing cells, (c) in nonexpressing cells, and (d) in CIP buffer adjusted to a pH of either 5 or 8.

interfering. As demonstrated, this is a potentially useful tool to follow the processing of such particles within the targeted cells. The integration of several functions in one system has some advantages over other existing systems, such as the possibility to study the degradation and processing in live-cell experiments without the need for additional cell-staining methods. In this study, CaCO3 particles were functionalized with a pHsensing fluorophore (i.e., FITC), followed by biocompatible and biodegradable polyelectrolyte coatings (i.e., PRM and DXS). As a reporter, GFP and DsRed encoding plasmid DNA was incorporated into the multilayers, allowing us to monitor the successful release and activity within the target cells. The FITClabeled core served two different purposes: (1) the fluorescence intensity was employed as sensor of the intracellular environment and (2) the fluorescence quenching with TB allowed us to discriminate between particles within cells and those attached to the outer cell membranes.

(1) De Geest, B. G.; De Koker, S.; Sukhorukov, G. B.; Kreft, I.; Parak, W. J.; Skirtach, A. G.; Demeester, J.; De Smedt, S. C.; Hennink, W. E. Soft Matter 2009, 5, 282–291. (2) Johnston, A. P. R.; Cortez, C.; Angelatos, A. S.; Caruso, F. Curr. Opin. Colloid Interface Sci. 2006, 11, 203–209. (3) Kim, B. S.; Choi, J. W. Biotechnol. Bioprocess Eng. 2007, 12, 323– 332. (4) Sukhorukov, G. B.; Moehwald, H. Trends Biotechnol. 2007, 25, 93– 98. (5) Kim, S.; Kim, J.-H.; Jeon, O.; Kwon, I. C.; Park, K. Eur. J. Pharm. Biopharm. 2009, 71, 420–430. (6) Petrov, A. I.; Volodkin, D. V.; Sukhorukov, G. B. Biotechnol. Prog. 2005, 21, 918–925. (7) Jewell, C. M.; Lynn, D. M. AdV. Drug DeliVery Syst. 2008, 60, 979– 999. (8) Selina, O. E.; Belov, S. Y.; Vlasova, N. N.; Balysheva, V. I.; Churin, A. I.; Bartkoviak, A.; Sukhorukov, G. B.; Markvicheva, E. A. Russ. J. Bioorg. Chem. 2009, 35, 103–110. (9) Kreft, O.; Javier, A. M.; Sukhorukov, G. B.; Parak, W. J. J. Mater. Chem. 2007, 17, 4471–4476. (10) Semmling, M.; Kreft, O.; Javier, A. M.; Sukhorukov, G. B.; Kaes, J.; Parak, W. J. Small 2008, 10, 1763–1768. (11) Fischlechner, M.; Zschoernig, O.; Hofmann, J.; Donath, E. Angew. Chem., Int. Ed. 2005, 44, 2892–2895. (12) Reddy, S. T.; Swartz, M. A.; Hubbell, J. A. Trends Immunol. 2006, 12, 573–579. (13) Cortez, C.; Tomaskovic-Crook, E.; Johnston, A. P. R.; Scott, A. M.; Nice, E. C.; Heath, J. K.; Caruso, F. ACS Nano 2007, 2, 93–102. (14) Berth, G.; Voigt, A.; Dautzenberg, H.; Donath, E.; Mo¨hwald, H. Biomacromolecules 2002, 3, 579–590. (15) Halozan, D.; Reibetanz, U.; Brumen, M.; Donath, E. Colloids Surf., A 2009, 342, 115–121. (16) Jin, Y.; Liu, W. C.; Wang, J. R.; Fang, J. H.; Gao, H. Colloids Surf., A 2009, 342, 40–45. (17) Reibetanz, U.; Halozan, D.; Brumen, M.; Donath, E. Biomacromolecules 2007, 6, 1927–1932. (18) Reibetanz, U.; Claus, C.; Typlt, E.; Hofmann, J.; Donath, E. Macromol. Biosci. 2006, 6, 153–160. (19) Volodkin, D. V.; Larionova, N. I.; Sukhorukov, G. B. Biomacromolecules 2004, 5, 1962–1972. (20) Nuutila, J.; Lilius, E. M. Cytometry, Part A 2005, 65, 93–102.

BM100237R