Cell-Based Drug Delivery Devices Using Phagocytosis-Resistant Backpacks

www.advmat.dewww.MaterialsViews.com

CO

MM

Cell-Based Drug Delivery Devices Using Phagocytosis-Resistant Backpacks
U
NIC

AT

Nishit Doshi , Albert J. Swiston , Jonathan B. Gilbert , Maria L. Alcaraz , Robert E. Cohen , Michael F. Rubner , * and Samir Mitragotri *
IO
N

Macrophages, ubiquitous phagocytic cells in the human immune system, play a key role in homeostatic, immunolo-gical, and infl ammatory processes. [ 1–2 ] Macrophages are widely distributed in various tissues and play a central role in clearing invading pathogens, dead cells, and foreign entities through phagocytosis. [ 3 ] Their wide presence in various organs and tis-sues makes them particularly suited to provide an immediate defense against invading threats. Moreover, macrophages are rapidly recruited to the diseased site by signaling molecules such as cytokines. Hence, macrophages are involved in a wide range of pathological conditions including cancer, atheroscle-rosis, various infl ammatory diseases such as vasculitis and asthma, and many others.

Since macrophages play an indispensable role in most patho-logical conditions, they represent an ideal target for therapeutic applications. Several approaches seeking to use macrophages for targeted therapies involve feeding therapeutic nanoparticles to macrophages ex vivo, followed by re-injection of the macro-phages to target the diseased site. This approach has shown promising results for treating HIV infections [ 4 ] , brain disor-ders, [ 5 ] and solid tumors. [ 6 ] While this strategy is effective for certain conditions, its applications are limited by the fact that the drug carriers are sequestered within the phagosome of macrophages, which reduces the release rates, and in certain cases, degrades the drug. This limitation can be potentially addressed by designing particles that: i) attach to the macro-phage surface, ii) avoid internalization, iii) do not interfere with macrophage function, and iv) release the encapsulated drugs in a controlled manner. However, development of materials that simultaneously fulfi ll these requirements is a signifi cant challenge. Herein, we report the ability of cellular backpacks to successfully encapsulate and controllably release drugs and

© 2011 WILEY-VCH Verlag GmAdv. Mater. 2011, 23, H105–H109

DOI: 10.1002/adma.201004074

N. Doshi , Prof. S. Mitragotri Department of Chemical EngineeringUniversity of California Santa BarbaraCA 93106, USA E-mail: [email protected] A. J. Swiston , Prof. M. F. Rubner Department of Materials Science and EngineeringMassachusetts Institute of TechnologyCambridge MA 02139, USAE-mail: [email protected] J. B. Gilbert , M. L. Alcaraz , Prof. R. E. Cohen Department of Chemical EngineeringMassachusetts Institute of TechnologyCambridge MA 02139, USA

avoid phagocytic internalization while remaining on the macro-phage’s surface. These characteristics point to new possibilities in creating cell-based bio-hybrid devices that leverage both the functions of the encapsulated cargo (drugs, nanoparticles, etc.) and the native functions of the cell.

Cellular backpacks are fabricated using a standard photoli-thography lift-off technique of layer-by-layer and spray deposited fi lm. [ 7 ] Briefl y, a positive photoresist is patterned with regularly spaced 7- μ m-diameter holes that extend down to the substrate. Next, a layer-by-layer deposited fi lm consisting of alternating hydrogen bond donor–acceptor pairs is deposited, and this layer comprises the release region that tethers the rest of the back-pack to the substrate. Two hydrogen-bonded regions were used, and details can be found elsewhere. [ 7 ] Next, a Polyelectrolyte multilayer (PEM) of either (FITC-PAH/MNP) or (PDAC/SPS) is deposited to provide suffi cient mechanical rigidity for the backpack to survive the fi nal acetone sonication step [FITC = fl ourescein, PAH = poly(allylamine hydrochloride), MNP = magnetic nanoparticles, PDAC = poly(diallyl dimethyl ammo-nium chloride), SPS = poly(styrene sulfonate)]. In the protein-eluting backpack construction, a PLGA/FITC-BSA fi lm is then sprayed onto the existing two layers and forms the hydrolytic payload region of the backpack [(PLGA = poly(lactic- co -glycolic acid), BSA = bovine serum albumin]. An optional second (PDAC/SPS) or (PAH/SPS) fi lm is deposited, and fi nally the cell-adhesive (HA/CHI; HA = hyaluronic acid, CHI = chitosan) region is added to the top of the heterostructure. A schematic of the three fi nal backpack fi lms is found in Figure 1 a (i–iii). The last fabrication step is sonicating the fi lm in acetone, which dis-solves the photoresist and simultaneously lifts off the deposited heterostructured fi lm in all areas except those adsorbed directly to the substrate. A SEM microscopy image of an as-fabricated backpack may be found in Figure 1 b.

Attachment and phagocytosis of backpacks was studied using J774 mouse macrophages as a model cell, and 6- μ m-diameter HA-coated spheres were used as control particles. Around 95% of backpacks added to the cells attached to the macrophage sur-faces ( n = 100 backpacks). Once attached, macrophages did not release backpacks from their surface, suggesting a strong attach-ment between the two. Time lapse video microscopy provided more insights into the cell-backpack interactions (Figure 1 c). Backpacks remained attached to macrophage surfaces without internalization even 50 min after introduction. In contrast, HA-coated spheres were phagocytosed within 30 min of attachment to the macrophage surface.

These qualitative observations were quantifi ed in terms of percent internalization. Particles were observed for 3 h fol-lowing attachment to the macrophage surface. At least 100

bH & Co. KGaA, Weinheim wileyonlinelibrary.com H105


CO

MM

UN

ICATI

ON

H106

Figure 1 . Backpacks for cell-based drug delivery devices: a) schematic image of the different backpack layers, all of which deposited on top of a photoresist-patterned glass substrate; b) scanning electron microscopy image of a backpack with a poly(methacrylic acid)/poly(vinylpyrrolidone) (PMAA/PVPON) release region fabricated by the above strategy. The average diameter is ≈ 6 μ m and the backpacks have a fl at disk shape. c) Time lapse microscopy image sequence of macrophages interacting with HA/CHI coated backpacks. The arrow indicates an internalized backpack. All other backpacks ( n = 7) are attached to macrophage surfaces for the entire time lapse sequence of 50 min and exhibit strong resistance to macrophage phagocytosis. d) Quantitative analysis of 6- μ m-diameter particles internalization by macrophages. The data are presented in terms of internalization percentage (percentage of particles observed that were internalized by phagocytosis). At least 100 particles were observed for each condition. Amine modifi ed spheres act as a positive control and show a high internalization percentage. HA coated backpacks show signifi cantly lower internalization percentage compared to both HA coated spheres and amine modifi ed spheres. e) Scanning electron micrographs showing the interaction of backpacks and spheres with macrophages after 3 h of incubation in standard cell culture conditions: i) HA coated backpack attached to the surface of a macro-phage, and ii) three 6- μ m-diameter HA coated spheres internalized within a macrophage.

particles were observed for each condition for statistical signifi -cance. Amine-modifi ed (positively charged at physiological pH) and HA-coated (negatively charged) spheres (diameter = 6 μ m) were used as positive controls. [ 8 ] As expected, amine-modifi ed particles exhibited high internalization with almost 80% of attached particles being phagocytosed. HA-coated spheres were

© 2011 WILEY-VCH Verlag Gmwileyonlinelibrary.com

internalized to a lesser extent (35% of attached particles); how-ever, a large proportion was still internalized. The difference between HA-coated and amine-modifi ed spheres likely origi-nates from a difference in their surface charge. HA was used for coating backpacks since interactions between HA and macro-phages are specifi c, mediated through the cell surface receptor

bH & Co. KGaA, Weinheim Adv. Mater. 2011, 23, H105–H109


CO

MM

UN

ICATIO

N

Figure 2 . Backpack attachment to macrophages does not affect cel-lular health and functions. a) Time-lapse video microscopy showing that a backpack-laden macrophage can still effi ciently internalize 3- μ m PS spheres (MP: macrophage, BP: backpack, PS-S: 3- μ m polystyrene sphere). The backpack is not internalized during the entire duration of the time lapse sequence. b) Differences in migration distance of macro-phages with and without backpacks: 3 h were found to be insignifi cant. c) MTT assay results in terms of proliferation potential, where 1 indicates healthy proliferation capacity and 0 indicates complete cell toxicity.

CD44. [ 9 ] Backpacks exhibited almost 30-fold lower extent of internalization compared to positive controls and about 14-fold lower extent of internalization compared to HA-coated spheres (Figure 1 d). It was found that backpacks typically attach to mac-rophages with their fl at face (Figure 1 e).

We assessed whether backpack attachment alters the cellular functions of macrophages. Two behaviors of macrophages with attached backpacks were assessed; i) the ability to internalize spher-ical particles (i.e., easily phagocytosed targets) and (ii) motility. We measured the internalization capability of macrophages with attached backpacks. PS spheres (diameter = 3 μ m) were chosen as the target since particles of this diameter have been shown to be preferentially phagocytosed by macrophages. [ 10 ] Macrophages were fi rst incubated with backpacks to allow suffi cient attachment and then 3- μ m PS spheres were added to the suspension. Time-lapse video microcopy clearly shows that backpack-laden cells were able to internalize 3- μ m spheres while simultaneously remaining attached to a backpack ( Figure 2 a). Macrophage motility was measured by tracking their center of mass for 3 h. No statistical difference was found between the mobility of macrophages with and without backpacks (Figure 2 b).

An MTT assay was performed to determine whether the attachment of backpacks adversely affects the health and pro-liferation of macrophages. The proliferation index of backpack-associated macrophages was not statistically different from the native population (Figure 2 c), indicating that backpacks are not acutely toxic to the cells and they do not interfere with a cell’s ability to reproduce. As a control, macrophages exposed to 6- μ m spheres also did not induce any noticeable toxicity.

The backpacks reported in this study have signifi cant bio-medical potential as multi-modal therapeutic and diagnostic platforms that may carry a variety of drugs, imaging/diag-nostic agents, or nanoparticles. Specifi cally, backpacks can be loaded with drugs which can be released in a controlled manner. As a proof of concept, we loaded the backpacks with fl uorescein conjugated bovine serum albumin (FITC-BSA) within a poly(lactic- co -glycolic acid) (PLGA) polymer matrix and measured subsequent protein release.

The fabrication method for drug releasing backpacks was slightly different from previously described due to the inclusion of PLGA ( Figure 3 a ), but the backpack still presented the same cell-adhesive and pH labile release regions as backpacks used in the toxicity and mobility studies. FITC-BSA was chosen since it is commonly used as a model protein and it could be replaced with a variety of therapeutic or diagnostic materials. We see that the backpacks released FITC-BSA in a burst manner with ≈ 40% released after 2 h and > 90% released after 24 h (Figure 3 b). In addition, it is worth noting that very similar release profi les were seen independent of the weight percentages of FITC-BSA within the polymer solution before spraying.

Macrophages are pervasive cells in the mammalian immune system, capable of identifying and neutralizing non-native agents found in the body. In addition, macrophages are routinely recruited to diseased sites, potentially making them highly effi -cient targeting devices. If non-native entities could be attached to these macrophages without being phagocytosed, these cells could serve as ideal chaperones for delivering drugs or other therapeutic tools to various pathological tissues. This is partic-ularly attractive for the treatment of cancer and infl ammatory


disorders, where macrophages are very strongly recruited. [ 11–13 ] This approach would offer much greater spatial resolution and selective uptake of therapeutic or diagnostic components, essentially using an individual cell as a cargo-carrying device.

Cellular backpacks are nanoscale-thickness microparticles fabricated using photolithography where polyelectrolyte multi-layers are assembled using the layer-by-layer technique, one of the most versatile methods to incorporate a variety of materials within fi lm structures for biochemical applications. [ 14,15 ] The



CO

MM

UN

ICATI

ON

H108

Figure 3 . Controlled release of FITC-BSA from therapeutic backpacks. a) Schematic image of the fabrication process. b) Controlled release pro-fi le of FITC-BSA in vitro. The normalized release of FITC-BSA occurred over multiple hours and was not dependent on the loading percentage of the spray emulsion used. Red and blue lines correspond to 7.5 wt% and 25 wt%, respectively.

backpacks are engineered to (1) contain therapeutic or diag-nostic materials, such as small molecule drugs, proteins, nano-particles, or functional polymers, and (2) attach to the surface of a cell using a non-toxic mechanism relying upon a natural ligand-receptor interaction such as the HA-CD44-mediated adhe-sion used in this study. [ 16,17 ] Because of their unique geometry

© 2011 WILEY-VCH Verlag Gwileyonlinelibrary.com

and capacity to contain therapeutic and diagnostic materials, cellular backpacks present unique opportunities as a functional phagocytosis-resistant microparticle.

Long-term immobilization of any particle on the surface of phagocytic cells like macrophages is extremely diffi cult. Recently, Stephan et al. showed the therapeutic potential of covalently attaching nanoparticles to the surface of cells, but phagocytic immature dendritic cells posed an internalization problem. [ 18 ] The primary challenge of attachment to phago-cytic cells, such as macrophages, is their propensity to engulf any material attached to their surface. Various properties of particles, particularly their size, surface chemistry, shape, and mechanical fl exibility have been shown to play a critical role in phagocytosis. [ 8 , 19–22 ] Particles in the range of 2–3 μ m are known to exhibit high attachment to macrophages, [ 19 , 23 ] and particles in this size range are very effi ciently phagocytosed. Although the rate of internalization depends on the size of the particle, most particles with 2–3- μ m diameters are internalized within 30 min of attachment to the macrophage surface.

Recent literature has shown that the local shape, orienta-tion, and mechanical properties of a particle signifi cantly infl u-ence its susceptibility to be internalized by phagocytosis. [ 20 ] Elongated, high-aspect-ratio particles and very fl at disk-shaped particles are shown to exhibit reduced phagocytosis compared to spherical particles. [ 24 ] The low internalization rate of disk-shaped backpacks reported here is consistent with these obser-vations. Although not considered in this study, the backpack’s fl exibility and low modulus may also contribute to low inter-nalization frequency, a phenomenon seen in previous work. [ 25 ] Indeed, the right blend of three factors (shape, orientation, and mechanical properties) mentioned, may make backpacks an ideal system for avoiding the phagocytosis behavior of macrophages.

Avoiding internalization by macrophages is a key feature of the backpack, but equally important is the observation that backpack attachment does not interfere with the macrophage’s native cellular functions. The results of the MTT proliferation assay suggest that backpacks are not toxic and do not affect the cell proliferation capacity. Furthermore, backpack attachment did not adversely affect a macrophage’s ability to internalize particles that are otherwise phagocytosable. Further investiga-tion is needed to understand whether or not backpack attach-ment affects other macrophage functions, including the release of chemokines and cytokines.

Cellular backpacks provide a unique opportunity for drug delivery and therapy since the polyelectrolyte multilayers that comprise much of the backpack are well studied for their bio-medical applications [ 26 ] and drug delivery properties. [ 27 ] In par-ticular, polyelectrolyte multilayers have shown the ability to deliver DNA, [ 28 ] vaccines, [ 29 ] proteins, [ 30 ] and small molecules. [ 31 ] The results presented here show that cellular backpacks are capable of delivering a model protein in a controlled and sus-tained manner in vitro. Another strategy for fi nely controlled drug release from therapeutic backpacks could be to trigger release by thermal ablation. Chemotherapy agents or particles capable of ablation via RF heating (i.e., gold or iron oxide) may be loaded into a backpack, which can then ride on a monocyte being recruited into a solid tumor. Using a recruited monocyte as an active therapeutic-carrying device may allow access to the

mbH & Co. KGaA, Weinheim Adv. Mater. 2011, 23, H105–H109


CO

MM

UN

ICATIO

N

relatively inaccessible hypoxic region of a tumor, [ 32 ] offering clear advantages over passive particle uptake through “leaky” vasculatures. Such a strategy has been adopted for gold nano-particles with promising results. [ 6 ]

Experimental Section Backpack Fabrication : The backpacks were formed through a

combination of photolithography and PEMs. Photolithographically patterned (PDAC4.0/SPS4.0) 15.5 -coated glass substrates were sequentially dipped in dilute polymer or nanoparticle solution using an automated Zeiss programmable slide stainer or nanoStrata dipping unit, the details for which have been previously described. [ 7 , 16 ] The fully charged polyelectrolyte and hydrogen-bonded release PEMs were built in the Zeiss dipper; the cell adhesive region was built in the nanoStrata unit. The formula for backpacks used in internalization studies is (PMAA3.0/PNIPAAM3.0) 80.5 (FITC-PAH3.0/MNP4.0) 10 (PAH3.0/SPS4.0) 30 , where the number following each species abbreviation indicates the solution pH and subscripts are the number of bilayers (half bilayer is indicated as 0.5; PNIPAAM = poly( N -isopropylacrylamide). An optional (CHI3.0/HA3.0) 3 region was used as the cell adhesive region for both backpacks and PS spheres. The purpose of each region is described in the backpack fabrication results section. See the Supporting Information for backpack compositions in toxicity and migration studies.

MTT Assay : J774 mouse macrophages were seeded at 10 4 cells/well onto a 96-well plate. Backpacks and 6- μ m spheres at a concentration of 10 5 per well (200 μ L media) were exposed to the cells for 4 h. Cells not exposed to any particles were used as negative control. Ten microliters of reagent from an MTT kit (Invitrogen, Carlsbad, CA) was applied to each well for 5 h. The liquid from each well was then replaced by DMSO and allowed to incubate in the dark at room temperature for about 10 min to lyse the cells. Absorbance was read at 570 nm ( λ max of MTT dye) and the proliferation index was calculated as the ratio of absorbance of the sample to the absorbance of negative control. The proliferation index values range from 0 to 1, with 0 indicating maximum mitochondrial toxicity and 1 representing high proliferation capacity and healthy cells.

Controlled Release of BSA from Backpacks : Backpacks for drug-release studies were slightly different in structure to enable the inclusion of a layer comprised of PLGA and FITC-BSA. The layer was applied by pressure spraying a dilute sonicated mixture of PLGA solubilized in chloroform and FITC-BSA solubilized in water (see Figure 3a and the Supporting Information for details). Backpacks were released from the surface using PBS pH 7.4. BSA release was performed in pH 7.4 PBS in static conditions at 37 ° C using fl uorescence analysis (see the Supporting Information for details). Prior to analysis, the sample was centrifuged at 10 000 relative centrifugal force and 0.8 mL of the supernatant was removed for analysis. Protein concentration was determined using a Fluorolog-3 spectrofl uorometer (Horiba Jobin Yvon, Japan) exciting at 494 nm and the emission was measured at 520 nm. Immediately following, 0.8 mL of fresh PBS was added to replace the removed solution. Data was measured in triplicate and each sample had one microscope slide of backpacks present. Approximately 13 cm 2 per slide were covered with backpacks with an approximate backpack density of 2 × 10 5 cm − 2 . After 5 days of elution the backpacks were isolated into 200 μ L of PBS, frozen, and lyophilized overnight. The samples were then dissolved in 250 μ L of dichloromethane and 200 μ L of DI water, and 800 μ L of PBS were added. The samples were sonicated for 90 min to increase the extraction surface area. Finally, the samples were centrifuged as before and the aqueous phase was removed for fl uorescence analysis.

Supporting Information Supporting Information is available from the Wiley Online Library or from the author.


Acknowledgements N.D.A. and A.J.S. contributed equally to this work. The authors acknowledge fi nancial support from the National Heart Lung and Blood Institute’s Program of Excellence in Nanotechnology (1UO1 HL080718) and a National Science Foundation Graduate Research Fellowship. This research was made with Government support under and awarded by DoD, Air Force Offi ce of Scientifi c Research, National Defense Science and Engineering Graduate (NDSEG) Fellowship, 32 CFR 168a. This research was supported in part by the U.S. Army through the Institute for Soldier Nanotechnologies, under Contract W911NF-07-D-0004 with the U.S. Army Research Offi ce and the MIT NSF funded MRSEC program (DMR 0819762).

Received: November 4, 2010 Revised: November 30, 2010

Published online: March 1, 2011

[ 1 ] B. Burke , C. Lewis , The Macrophage , Oxford Univeristy Press , USA , 2002 . [ 2 ] A. Lasser , Hum. Pathol. 1983 , 14 , 108 . [ 3 ] A. Aderem , D. Underhill , Annu. Rev. Immunol. 1999 , 17 , 593 . [ 4 ] H. Dou , C. J. Destache , J. R. Morehead , R. L. Mosley , M. D. Boska ,

J. Kingsley , S. Gorantla , L. Poluektova , J. A. Nelson , M. Chaubal , J. Werling , J. Kipp , B. E. Rabinow , H. E. Gendelman , Blood 2006 , 108 , 2827 .

[ 5 ] H. Dou , C. B. Grotepas , J. M. McMillan , C. J. Destache , M. Chaubal , J. Werling , J. Kipp , B. Rabinow , H. E. Gendelman , J Immunol 2009 , 183 , 661 .

[ 6 ] M. R. Choi , K. J. Stanton-Maxey , J. K. Stanley , C. S. Levin , R. Bardhan , D. Akin , S. Badve , J. Sturgis , J. P. Robinson , R. Bashir , N. J. Halas , S. E. Clare , Nano Lett. 2007 , 7 , 3759 .

[ 7 ] A. Swiston , C. Cheng , S. Um , D. Irvine , R. Cohen , M. Rubner , Nano Lett. 2008 , 8 , 4446 .

[ 8 ] L. Thiele , H. Merkle , E. Walter , Pharm. Res. 2003 , 20 , 221 . [ 9 ] R. Peach , D. Hollenbaugh , I. Stamenkovic , A. Aruffo , J. Cell Biol.

1993 , 122 , 257 . [ 10 ] Y. Tabata , Y. Ikada , Biomaterials 1988 , 9 , 356 . [ 11 ] A. Luster , N. Engl. J. Med. 1998 , 338 , 436 . [ 12 ] G. Hansson , N. Engl. J. Med. 2005 , 352 , 1685 . [ 13 ] L. Coussens , Z. Werb , Nature 2002 , 420 , 860 . [ 14 ] K. Ariga , J. Hill , Q. Ji , Phys. Chem. Chem. Phys. 2007 , 9 , 2319 . [ 15 ] F. Caruso , Adv. Mater. 2001 , 13 , 11 . [ 16 ] A. Swiston , J. Gilbert , D. Irvine , R. Cohen , M. Rubner , Biomacromol-

ecules 2010 , 11 , 4920 . [ 17 ] K. Goa , P. Benfi eld , Drugs 1994 , 47 , 536 . [ 18 ] M. Stephan , J. Moon , S. Um , A. Bershteyn , D. Irvine , Nat. Med.

2010 , 16 , 1035 . [ 19 ] J. Champion , A. Walker , S. Mitragotri , Pharm. Res. 2008 , 25 , 1815 . [ 20 ] J. Champion , S. Mitragotri , Proc. Natl. Acad. Sc. USA 2006 , 103 , 4930 . [ 21 ] S. Mitragotri , J. Lahann , Nat. Mater. 2009 , 8 , 15 . [ 22 ] G. Martínez , N. Csaba , S. Fischer , A. Sichelstiel , T. Kündig ,

B. Gander , P. Johansen , J. Controlled Release 2008 , 130 , 161 . [ 23 ] N. Doshi , S. Mitragotri , PLoS One 2010 , 5 , DOI:10.1371/journal.

pone.0010051. [ 24 ] J. Champion , S. Mitragotri , Pharm. Res. 2009 , 26 , 244 . [ 25 ] K. Beningo , Y. Wang , J. Cell Sci. 2002 , 115 , 849 . [ 26 ] Z. Y. Tang , Y. Wang , P. Podsiadlo , N. A. Kotov , Adv. Mater. 2006 , 18 , 3203 . [ 27 ] A. N. Zelikin , ACS Nano , 4 , 2494 . [ 28 ] C. M. Jewell , D. M. Lynn , Adv. Drug Deliv. Rev. 2008 , 60 , 979 . [ 29 ] X. F. Su , B. S. Kim , S. R. Kim , P. T. Hammond , D. J. Irvine , ACS

Nano 2009 , 3 , 3719 . [ 30 ] M. Macdonald , N. M. Rodriguez , R. Smith , P. T. Hammond , J. Con-

trolled Release 2008 , 131 , 228 . [ 31 ] B. S. Kim , R. C. Smith , Z. Poon , P. T. Hammond , Langmuir 2009 , 25 , 14086 . [ 32 ] A. Primeau , A. Rendon , D. Hedley , L. Lilge , I. Tannock , Clin. Cancer

Res. 2005 , 11 , 8782 .


Documents

Cell-Based Drug Delivery Devices Using Phagocytosis-Resistant Backpacks