Nanomedicine: Nanotechnology, Biology, and Medicine11 (2015) 1861–1869

Original Article

Thermo-responsive polymeric nanoparticles for enhancing neuronaldifferentiation of human induced pluripotent stem cells

Hye In Seo, BSa,1, Ann-Na Cho, BSb,1, Jiho Jang, Ph.Dc, Dong-Wook Kim, Ph.Dc,Seung-Woo Cho, Ph.Db,d,⁎, Bong Geun Chung, Ph.Da,⁎⁎

aDepartment of Mechanical Engineering, Sogang University, Seoul, Republic of KoreabDepartment of Biotechnology, Yonsei University, Seoul, Republic of Korea

cDepartment of Physiology, Yonsei University College of Medicine, Seoul, Republic of KoreadDepartment of Neurosurgery, Yonsei University College of Medicine, Seoul, Republic of Korea

Received 20 February 2015; accepted 25 May 2015

nanomedjournal.com

Abstract

We report thermo-responsive retinoic acid (RA)-loaded poly(N-isopropylacrylamide)-co-acrylamide (PNIPAM-co-Am) nanoparticles fordirecting human induced pluripotent stem cell (hiPSC) fate. Fourier transform infrared spectroscopy and 1H nuclear magnetic resonanceanalysis confirmed that RA was efficiently incorporated into PNIAPM-co-Am nanoparticles (PCANs). The size of PCANs dropped withincreasing temperatures (300-400 nm at room temperature, 80-90 nm at 37 °C) due to its phase transition from hydrophilic to hydrophobic.Due to particle shrinkage caused by this thermo-responsive property of PCANs, RA could be released from nanoparticles in the cells uponcellular uptake. Immunocytochemistry and quantitative real-time polymerase chain reaction analysis demonstrated that neuronaldifferentiation of hiPSC-derived neuronal precursors was enhanced after treatment with 1-2 μg/ml RA-loaded PCANs. Therefore, wepropose that this PCAN could be a potentially powerful carrier for effective RA delivery to direct hiPSC fate to neuronal lineage.

From the Clinical Editor: The use of induced pluripotent stem cells (iPSCs) has been at the forefront of research in the field of regenerativemedicine, as these cells have the potential to differentiate into various terminal cell types. In this article, the authors utilized a thermo-responsivepolymer, Poly(N-isopropylacrylamide) (PNIPAM), as a delivery platform for retinoic acid. It was shown that neuronal differentiation could beenhanced in hiPSC-derived neuronal precursor cells. This method may pave a way for future treatment of neuronal diseases.© 2015 Elsevier Inc. All rights reserved.

Key words: Thermo-responsive nanoparticle; Poly(N-isopropylacrylamide)-co-acrylamide;Retinoic acid;Human induced pluripotent stemcells;Neuronal differentiation

Induced pluripotent stem cells (iPSCs) possess great potentialfor therapeutic applications involving cell replacement. Epige-

This research was supported by a grant from the Korea Health TechnologyR&DProject through theKoreaHealth IndustryDevelopment Institute, funded bytheMinistry ofHealth&Welfare, Republic ofKorea (Grant numberHI14C3347);by the Bio & Medical Technology Development Program of the NationalResearch Foundation (NRF) funded by the Ministry of Science, ICT & FuturePlanning (2015M3A9D7030461); by a grant (NRF-2010-0020409) from theNational Research Foundation of Korea (NRF), funded by theMSIP, Republic ofKorea; and in part by a grant (HI14C1588) from the Korea Health TechnologyR&D Project, funded by the Ministry of Health and Welfare, Republic of Korea.⁎ Correspondence to: S-WCho,DepartmentofBiotechnology,YonseiUniversity,

Seoul, Korea.⁎⁎Correspondence to: B.G. Chung, Department of Mechanical Engineering,Sogang University, Seoul, Korea.

E-mail addresses: seungwoocho@yonsei.ac.kr (S.-W. Cho),bchung@sogang.ac.kr (B.G. Chung).1 These authors contributed equally to this work.

Please cite this article as: Seo HI, et al, Thermo-responsive polymeric nanoparstem cells. Nanomedicine: NBM 2015;11:1861-1869, http://dx.doi.org/10.1016

http://dx.doi.org/10.1016/j.nano.2015.05.0081549-9634/© 2015 Elsevier Inc. All rights reserved.

netically reprogrammed iPSCs have been found to haveself-renewal and pluripotent capacity similar to embryonicstem cells.1,2 Recently, iPSC-derived neurogenesis has beenstudied for applications in the research of neurodegenerativediseases (e.g., Parkinson’s disease, Alzheimer’s disease, andischemic stroke) and neurological disorders. Diverse approachessuch as genetic modifications (viral vectors, mesoporousnanoparticles),3,4 co-culture (with astroglial cells),5 and biore-actors (rotary suspension system)6 have been developed topromote neuronal differentiation of iPSCs for the purpose oftherapeutic applications and neuronal disease modeling. Inparticular, the delivery of transcription factors regulatingneuronal differentiation holds great potential for the develop-ment of new populations of neuronal cell lineages from iPSCs.

Among such transcription factors, retinoic acid (RA) has drawnattention for its capacity for neuronal differentiation of stem cells.Ametabolic compound derived fromvitaminA,RA iswell-knownto be involved in neural differentiation,7 axon outgrowth,8 and

ticles for enhancing neuronal differentiation of human induced pluripotent/j.nano.2015.05.008

1862 H.I. Seo et al / Nanomedicine: Nanotechnology, Biology, and Medicine 11 (2015) 1861–1869

patterning of the neural plate and the neural tube.9 For thosereasons, RA has been the most commonly used morphogen toinduce neurogenesis.10–16 However, RA is unstable and rapidlydegraded under physiological conditions. In addition, the solubilityof RA is low in the aqueous phase.17 Therefore, nanoparticleformulations facilitating stability, solubility, and cellular uptake ofRA have been sought for stem cell engineering. For example, aRA-loaded polyethylenimine (PEI) nanoparticle formulationwas previously developed to enhance neuronal differentiationof neural stem cells or embryonic stem cells.10,18,19

Thermo-responsive polymeric nanomaterials produce greatbenefits for stem cell engineering. Poly(N-isopropylacrylamide)(PNIPAM), one of the representative thermo-responsive polymers,has been applied in various forms of nanogels, injectable hydrogels,particles, and surface coating films for regulating stem celldifferentiation and expansion.20–26 Given its thermo-responsiveproperties, PNIPAM-based biomaterials could be especially usefulas delivery platforms to direct stem cell fate. For instance, theself-assembled PNIPAM-co-acrylic acid (AA) nanogel hasbeen used as a gene delivery carrier for human mesenchymalstem cells.20 PNIPAM-co-AA nanogel encapsulated with PEIwas internalized into human mesenchymal stem cells to releaseplasmid DNA within the cytosol,20 suggesting that it couldefficiently deliver specific genes into stem cells.27,28 It has alsobeen reported that thermo-responsive PNIPAM-based deliverysystems can control the release of several growth factors.22,25,29

One of the interesting features of PNIPAM is its lower criticalsolution temperature (LCST) in aqueous solutions.30,31 BecausePNIPAM chains undergo rapid structural changes like coil-to-globule transition above the LCST,32–34 PNIPAM particlesabruptly shrink.35,36 Therefore, hydrophobic drugs such as RAcan be temporally released from PNIPAM-based biomaterialsabove the LCST. Although previous approaches have investi-gated stem cell differentiation using PNIPAM-based biomate-rials, there are no studies to date that have tested the delivery ofRA using thermo-responsive PNIPAM-based nanoparticles foriPSC differentiation.37

Herein, we first report the efficacy of RA-loaded thermo-responsive PNIPAM-based nanoparticles for directing neuronallineage differentiation from hiPSCs. Thermo-responsive PNI-PAM-co-acrylamide (PNIPAM-co-Am) nanoparticles were syn-thesized for efficient intracellular delivery of RA and thenapplied to neuronal precursor cells derived from hiPSCs. Ourresults demonstrate that RA-loaded PNIPAM-co-Am nanoparti-cles (PCANs) significantly enhance neuronal differentiation ofhiPSC-derived neuronal precursor cells.

Methods

Synthesis of PCANs

To synthesize PCANs, 500 mg NIPAM (Sigma-Aldrich, St.Louis, MO, USA), 13.5 mg N,N′-Methylenebisacrylamide(MBA, Sigma-Aldrich), 100 mg sodium dodecyl sulfate (SDS,Sigma-Aldrich), and 12.5 mg Am (Sigma-Aldrich) were addedto 50 ml distilled water. The solution was mixed with nitrogengas at room temperature for 20 minutes. Potassium persulfate

(37.5 mg, KPS, Dae Jung, Siheung, Gyonggi, Korea) was addedto initiate polymerization. The solution was then allowed to reactwith nitrogen gas at 70 °C for 5 hours. The polymerizationsolution was dialyzed against distilled water for 7 days toremove unreacted monomers. The distilled water was replen-ished every day during dialysis. The polymerization solution wassubsequently lyophilized.

Preparation of RA-loaded PCANs

PCANs were dissolved in distilled water (10 mg/ml). RA(Sigma-Aldrich) was dissolved in dimethyl sulfoxide (DMSO,Sigma-Aldrich) to prepare a stock solution (5 mg/ml). The RAsolution was subsequently added drop wise to the PCANsolution. The combined solutions were stirred at roomtemperature in the dark for 24 hours. The RA-loaded PCANsolution was dialyzed against distilled water for 24 hours using adialysis membrane (MWCO 6,000-8,000, Spectrum Laborato-ries, Inc., CA, USA) to remove unreacted RA. After dialysis,RA-loaded PCAN solutions were lyophilized (FDU-1200,Tokyo Rikakikai Co., LTD., Miyagi, Japan) at -50 °C for 4 days.

Preparation of fluorescently labeled RA-loaded PCANs

RA-loaded PCANs were dissolved in distilled water. Fluores-cein isothiocyanate-dextran solution (FITC-dextran, Sigma-Aldrich) was added drop wise to the nanoparticle solution. Thesolution was stirred at room temperature in the dark for 12 hours.The nanoparticle solution was dialyzed against distilled water for24 hours using a dialysis membrane. After dialysis, the nanopar-ticle solution was lyophilized to obtain a powder.

Transmission electronic microscopy (TEM)

PCANs were dispersed in distilled water using a sonicator(Vibra-Cell, Sonics, Newtown, CT, USA), while being chilled inan ice bath. The dispersion was applied to copper grids and left toevaporate at room temperature. The prepared samples werescanned using a TEM (JEM 1010, JEOL Ltd., Tokyo, Japan) at avoltage of 80 kV.

Fourier transmittance infrared (FTIR) spectra

PCANs and RA-loaded PCANs were dispersed in ethanol.The dispersion was dropped into the sample holder of the FTIRspectrometer. The solvent was subsequently left to evaporate atroom temperature. FTIR spectra of PCANs and RA-loadedPCANs were recorded using an FTIR spectrometer (NICOLETAVATAR 330, Thermo Nicolet, Madison, WI, USA).

1H nuclear magnetic resonance (NMR) spectra

PCANs and RA-loaded PCANs were dispersed in DMSO-d6.1H NMR spectroscopy of nanoparticles was analyzed using aUnity-Inova instrument (Agilent Technologies, Santa Clara, CA,USA) at 500 MHz.

Dynamic light scattering (DLS)

Lyophilized PCAN powder was dispersed in distilled waterand sonicated in an ice bath. The dispersion, placed in a DLScuvette, was heated at a rate of 1 °C/min in the temperature range

Figure 1. Synthesis and thermo-responsive property of PCANs. (A) Synthesis of PCANs by radical polymerization. (B) TEM image of PCANs. Scale bar is200 nm. (C) LCST and deswelling ratio (D) of PCANs with respect to temperature change. (E) Schematic illustration of RA release from PCANs. (F) Sizedistribution of PCANs measured by DLS analysis at 37 °C (n = 4).

1863H.I. Seo et al / Nanomedicine: Nanotechnology, Biology, and Medicine 11 (2015) 1861–1869

of 20–50 °C. The dispersion was allowed to equilibrate for120 seconds before each measurement. The LCST of PCANswas measured by using a DLS analyzer (Zetasizer Nano ZS,Malvern Instruments Ltd., Malvern, UK). The deswelling ratio(α) of PCANs was calculated using the following equation:

α ¼ d=d0ð Þ3

where d0 is the diameter of PCANs at 20 °C and d is the diameterof PCANs at 20–50 °C.

The size distribution of PCANs was also determined by usinga DLS analyzer at 37 °C (n = 4). To measure the zeta potential,the PCANs and RA-loaded PCANs were dispersed in PBS atpH 7.4. The nanoparticle dispersion inserted into zeta potentialcuvette. The zeta potential of PCANs and RA-loaded PCANswas measured by a DLS analyzer.

Human induced pluripotent stem cell (hiPSC) culture anddifferentiation

HiPSCs (cell line; J10) were provided from Yonsei UniversityCollege of Medicine. The cells were cultured in hiPSC medium[Dulbecco’s Modified Eagle’s Medium/Nutrient Mixture F12

(DMEM/F12, Invitrogen, Carlsbad, CA, USA), 20% knockout-serum replacement (Invitrogen), 1% penicillin-streptomycin (Invi-trogen), 1 × non-essential amino acid (Invitrogen), 0.1 mM β-mercaptoethanol (Sigma), and 10 ng/ml basic fibroblast growthfactor (bFGF) (Peprotech, Rocky Hill, NJ, USA)].38 Undifferenti-ated hiPSCs were maintained on feeder cell layers made of STOfibroblasts (American Type Culture Collection (ATCC), Manassas,VA, USA) inactivated by mitomycin C (10 μg/ml, Sigma-Aldrich).For hiPSC differentiation, an embryoid body (EB) was firstgenerated by culturing hiPSCs for 4 days in non-adherent Petridishes using hiPSC medium without any bFGF. To inducedifferentiation of hiPSCs into neuronal precursor cells, 5 μMdorsomorphin (DM) (Sigma-Aldrich) and 5 μM SB431542(Sigma-Aldrich) were added to the hiPSC medium during EBformation. EBs were then attached to the culture dish coated withMatrigel (BD Biosciences, San Jose, CA, USA) and were culturedin neural induction medium [DMEM/F12 medium (Invitrogen)supplemented with 1 × N2 supplement (Invitrogen) and 1 × non-essential amino acids (Invitrogen)].38,39 After culturing for anadditional 4–5 days, neural rosettes appeared in the center of theEBs. Neural crest cells are usually localized in the periphery of theattached EBs.38,40

Figure 2. Analysis of FTIR spectrum of RA, PCAN, and RA-loaded PCAN.

1864 H.I. Seo et al / Nanomedicine: Nanotechnology, Biology, and Medicine 11 (2015) 1861–1869

Treatment with RA-loaded PCANs

Neural rosettes were isolated from the surrounding neuralcrest cells using modified Pasteur pipettes with the end of ahook-like shape. The collected neural rosettes were mechanicallydissociated into small clumps, re-seeded onto Matrigel-coatedculture dishes, and submerged in neural induction medium(3.0 × 105 cells/ml). The neural rosette clumps were allowed tostabilize overnight and were then treated with RA-loadedPCANs at different concentrations (1, 2, 5, and 10 μg/ml). Onthe next day, the medium was replaced with fresh neuralinduction medium. Four days after treatment with RA-loadedPCANs, the cytotoxicity of hiPSC-derived neural rosettes wasexamined using the Live/Dead Cell Viability assay kit (Invitro-gen). The cell viability was quantified from the Live/Dead-stained fluorescent images by using Image J program (NationalInstitutes of Health, Bethesda, MD, USA). Cell viability waspresented as the percentage ratio of green-positive live cells tototal cell population. Neuronal marker expression of hiPSC-derived neural rosettes was examined 9 days after treatment withRA-loaded PCANs.

Cellular uptake of RA-loaded PCANs

Cellular uptake of PCANs into hiPSC-derived neural rosetteswas visualized using a confocal microscope (LSM 700, CarlZeiss, Jena, Germany). Neural rosette clumps were seeded ontoMatrigel-coated culture dishes and left to stabilize for 1 day.Then, the cells were treated with FITC-labeled, RA-loadedPCANs (2 μg/ml). After 18 hours of incubation, the cells werefixed in a 4% paraformaldehyde (PFA) (Sigma-Aldrich) solutionfor 20 minutes at room temperature and subsequently imagedwith a confocal microscope. Cell nuclei were counterstained with4′,6-diamidino-2-phenylindole (DAPI) (Tokyo Chemical Indus-try, Tokyo, Japan).

Immunocytochemistry

Immunocytochemical staining was performed to examineneuronal differentiation of neural rosettes treated withRA-loaded PCANs as previously reported.41 The cells werefixed in a 4% PFA solution for 20 minutes at room temperatureand permeabilized with 0.1% (v/v) Triton X-100 (Sigma-Aldrich) for 5 minutes. After washing with phosphate-bufferedsaline (PBS), the cells were incubated in 2% (w/v) bovine serumalbumin (Wako Pure Chemical Industries, Osaka, Japan)solution for 30 minutes. The cells were then incubated withprimary antibodies at 4 °C. The following primary antibodieswere used: mouse monoclonal anti-neuron-specific class IIIβ-tubulin (Tuj1) (1:200; Santa Cruz Biotechnology, Santa Cruz,CA, USA) and rabbit polyclonal anti-neuron microtubule-associated protein family (MAP2) (1:200; Santa Cruz Biotech-nology). After overnight incubation of primary antibodies, thecells were washed with PBS and incubated with secondaryantibodies [Alexa Fluor-488 goat anti-mouse IgG (1:200;Invitrogen) and Alexa Fluor-594 donkey anti-rabbit IgG(1:200; Invitrogen)] for 1 hour at room temperature. Cell nucleiwere counterstained with DAPI. Images of the stained cells weretaken using a fluorescence microscope (1X 71, Olympus, Tokyo,Japan). The density of MAP2-positive cells (per image) in eachgroup was quantified from MAP2-stained images.

Quantitative real-time polymerase chain reaction (qPCR)

The expression of neuronal markers in hiPSC-derived neuralrosettes was investigated using qPCR analysis as previouslydescribed.41 Total RNA was isolated from the cells using anRNeasy Mini kit (Qiagen, Chatsworth, CA, USA) following themanufacturer’s instructions. RNA concentration was determinedby measuring the absorbance of the samples at 260 nm using aspectrophotometer (Infinite M200 Pro, Tecan, Maennedorf,Switzerland). Total RNA was reverse-transcribed into cDNAusing a TaKaRa PrimeScript II First strand cDNA synthesis kit(TaKaRa, Shiga, Japan). qPCR analysis was performed using theStepOnePlus Real-Time PCR system (Applied Biosystems,Foster City, CA, USA). The synthesized cDNA and the TaqManFast Universal PCRMaster Mix (Applied Biosystems) were usedfor the PCR reaction. The gene expression was quantified withthe TaqMan Gene Expression Assay (Applied Biosystems) forthe following targets: Tuj1 (Hs00801390_s1) and MAP2(Hs00258900_m1). The relative expression level of each targetgene was determined using the comparative Ct method andnormalized to the endogenous reference (glyceraldehyde3-phosphate dehydrogenase (GAPDH): Hs02758991_g1).42

Statistical analysis

Quantitative data are expressed as means ± standard devia-tions. Statistical significance between experimental groups andcontrol groups was determined as previously described.43

Statistical analysis was performed using the unpaired Student'st-test using GraphPad Prism 6 software (Graphpad Software, SanDiego, CA, USA). A P-value smaller than 0.05 was consideredstatistically significant.

Figure 3. Analysis of 1H-NMR spectrum of (A) PCAN and (B) RA-loaded PCAN.

1865H.I. Seo et al / Nanomedicine: Nanotechnology, Biology, and Medicine 11 (2015) 1861–1869

Results

Synthesis and analysis of RA-loaded PCANs

PCANs were synthesized through a radical polymerizationreaction (Figure 1, A). TEM images revealed the sphericalmorphology of PCANs (Figure 1, B). DLS analysis showed thatthe diameter of PCANs was approximately 300-400 nm at roomtemperature. When the temperature was raised above the LCST(35 °C), the diameter of the nanoparticles shrank to 80-90 nm(Figure 1, C). The size distribution of PCANs determined by DLSanalysis indicated that D10, D50, and D90 were 43.49 ±0.41 nm,65.09 ± 1.27 nm, and 98.38 ± 3.44 nm, respectively (Figure 1, F).The analysis of the deswelling ratio of PCANs in the range of20-50 °C indicated that PCANs deswelled above 35 °C LCST(Figure 1, D). Deswelling of PCANs leads to RA release fromPCANs in a temperature-dependent manner (Figure 1, E). We alsosynthesized RA-loaded PCANs. RA was loaded into PCANs viahydrophobic interactions and electrostatic bonding. To analyzewhether RA was incorporated into PCANs, we used FTIR and 1HNMR analysis. With FTIR analysis the NH stretch of the 1st and 2nd

amide was observed at 3363 and 3288 cm-1 (Figure 2). The C = Ostretch of the 1st and 2nd amide was found at 1635 cm-1. The NHbend of the 1st and 2nd amide was observed at 1558 cm-1. TheFTIR results therefore confirmed the synthesis of PCANs. Theyalso confirmed that RA was incorporated into PCANs by detectingthe vinyl out-of-plane (oop) of RA at 1045 cm-1 (Figure 2).Furthermore, 1H NMR analysis showed the proton peak (m) of theamide group, the methine proton peak (n) of isopropyl, and theproton peak (o) of the methyl group of isopropyl (Figure 3, A). Wealso observed the proton peak (a) of carboxyl acid from the RA,demonstrating that RA was incorporated into PCANs (Figure 3, B).To investigate the surface charge, we measured the zeta potential of

PCANs and RA-loaded PCANs. It represented that the zetapotential of PCANs and RA-loaded PCANs in PBS at pH 7.4 was−4.70 ± 0.3 and -3.81 ± 0.8 mV, respectively.

Cytotoxicity and cellular uptake of RA-loaded PCANs

Cytotoxicity was examined at different doses of RA-loadedPCANs (1, 2, 5, and 10 μg/ml). After 4 days of treatment, neuralrosettes cultured with nanoparticle doses of 1, 2, and 5 μg/ml werehighly viable, indicating no significant cytotoxicity of nanoparti-cles at these doses compared to the control group (no treatment)(Figure 4, A-D). However, at a concentration of 10 μg/ml,nanoparticles exhibited cytotoxicity as indicated by a relativelylarger number of dead cells (Figure 4, E). The viability of the cellstreated with 1 μg/ml and 2 μg/ml RA-loaded PCANs (about 80%)was not significantly different (P N 0.05) from that of the cells inno treatment group (Figure 4, F). The increase in the concentrationof RA-loaded PCANs to 10 μg/ml led to a significant decrease incell viability. Next, cellular uptake was investigated withFITC-labeled RA-loaded PCANs (2 μg/ml). Confocal microscopicobservation revealed that FITC signals could be detected in thecytoplasm of hiPSC-derived neural rosettes 18 hours afternanoparticle treatment (Figure 5), indicating efficient cellularuptake of PCANs. We conclude that the positive charge of theamine group in the PCANs contributed to enhance cellular uptakeof RA-loaded nanoparticles into hiPSC-derived neural rosettes.

Enhanced neuronal differentiation of hiPSCs cultured with RA-loaded PCANs

To promote neuronal differentiation of hiPSCs, hiPSC-derivedneural rosettes were treated with RA-loaded PCANs. After 9 days,the differentiated cells derived from neural rosettes were

Figure 4. The viability of hiPSC-derived neural rosettes 4 days after treatment of RA-loaded PCANs. Live/Dead staining assay was performed to examine theviability of cells in each group. Live cells stains green and dead cells stains red. (A) Control group without treatment of RA-loaded PCANs. (B) 1 μg/ml, (C)2 μg/ml, (D) 5 μg/ml, and (E) 10 μg/ml of RA-loaded PCANs. Scale bar is 200 μm. (F) The viability of cells treated with RA-loaded PCANs determined fromthe Live/Dead-stained fluorescent images (n = 4, **; P b 0.01 versus no treatment group).

1866 H.I. Seo et al / Nanomedicine: Nanotechnology, Biology, and Medicine 11 (2015) 1861–1869

characterized using immunocytochemistry and qPCR analysis fortwo neuronal markers (Tuj1 and MAP2). Immunocytochemicalstaining revealed that Tuj1 and MAP2 were strongly expressed inhiPSC-derived neural rosettes treated with 1, 2, and 5 μg/mlRA-loaded PCANs (Figure 6, A-D). Interestingly, nanoparticletreatment at these concentrations generated a highly extendedmorphology in MAP2-positive cells (Figure 6, A-D) and asignificantly increased number of MAP2-positive cells (Figure 6, E)compared to the control group. Similarly, qPCR analysisdemonstrated that gene expression of the Tuj1 marker was muchhigher in the cells treated with 1 and 2 μg/ml RA-loaded PCANsthan in control cells (Figure 6, F). The expression of MAP2 wasalso increased in neural rosettes treated with 1, 2, and 5 μg/mlRA-loaded PCANs (Figure 6, G). These results indicate thatPCANs transferred into hiPSC-derived neural rosettes can releaseRA into the cells, promoting neuronal differentiation of neuronalprecursor cells derived from hiPSCs due to the thermo-responsiveproperties of PCANs (Figure 1, E).

Discussion

The thermo-responsive property is one of the major featuresof RA-loaded PCANs. At temperatures above the LCST, thehydrogen bonds between the amide groups of PCANs and water

molecules were broken, and the coil-to-globule transition began.This caused the phase transition of the PCANs from hydrophilicto hydrophobic. This phase transition induced a deswelling ofPCANs (Figure 1, D), leading to RA release from PCANs in atemperature-dependent manner (Figure 1, E). The surface chargeof nanoparticles is of great importance for cellular uptakeapplications. The nanoparticle with negative charge might berepulsive to cell membranes with the large negatively-chargeddomains. Thus, it has been known that nanoparticles withnegative charge could bind to cationic sites on the surface of thecells in the form of nanoparticle clusters.44,45 Although PCANsand RA-loaded PCANs showed negative charge, we demon-strated the efficient cellular uptake of PCANs and RA-loadedPCANs. It was probably because PCANs with negative chargecould adhere to cationic sites on the cell surfaces due to therepulsive interaction with negatively-charged portion of cellmembranes and subsequently undergo endocytosis-mediatedcellular uptake as previously described.46

RA is a potent transcription factor for enhancing neurogen-esis, but formulation and delivery of RA are quite challengingdue to the poor stability of RA in physiological conditions and itslow solubility in aqueous solutions.17 Therefore, severalapproaches using nanoparticles have aimed to overcome theseinherent limitations by avoiding cellular degradation andfacilitating efficient intracellular delivery of RA.18,19,47,48 Maia

Figure 5. Cellular uptake of FITC-labeled, RA-loaded PCANs into hiPSC-derived neural rosettes 18 hours after nanoparticle treatment. (A) Control groupwithout FITC-RA-PCAN treatment. (B) The cells treated with FITC-RA-PCANs. Scale bar is 200 μm.

1867H.I. Seo et al / Nanomedicine: Nanotechnology, Biology, and Medicine 11 (2015) 1861–1869

et al. demonstrated that polyelectrolyte nanoparticles composedof RA, PEI, and dextran sulfate (for efficient internalization ofRA) can induce neuronal differentiation of neural stem cells.18

Figure 6. Immunofluorescent staining and qPCR analysis. Immunofluorescent stai9 days after treatment of RA-loaded PCANs. (A) Control group without treatmRA-loaded PCANs. (E) The density of MAP2-positive cells (n = 8–9, **; P bdetermine the expression of the neuronal markers (F) Tuj1 and (G) MAP2 in hiP(9 days after the treatments) (n = 3, *; P b 0.05 versus no treatment group). The gthat of control group without nanoparticle treatment.

The same formulation was successfully applied to controlneurogenesis of neural stem cells from the subventricular zonein the brain.10 We have previously reported that RA-loaded PEI

ning of hiPSC-derived neural rosettes for neuronal markers (Tuj1 and MAP2)ent of RA-loaded PCANs, (B) 1 μg/ml, (C) 2 μg/ml, and (D) 5 μg/ml of0.01 versus no treatment group). Scale bar is 200 μm. qPCR analysis toSC-derived neural rosettes with or without treatment of RA-loaded PCANsene expression of the cells treated with RA-loaded PCANs was normalized to

1868 H.I. Seo et al / Nanomedicine: Nanotechnology, Biology, and Medicine 11 (2015) 1861–1869

nanoparticles induce embryonic stem cell-derived neuronaldifferentiation.191H NMR analysis demonstrated that cationichydrophilic PEI conjugated into anionic hydrophobic RA.19

Immunocytochemistry analysis showed that RA released fromPEI nanoparticles can induce differentiation of murine embryonicstem cells into Tuj1-positive, neural-like cells.19 In our currentstudy, PCANs were formulated for efficient intracellular deliveryof RA, which significantly enhanced neuronal differentiation ofhiPSCs into Tuj1- and MAP2-positive cells (Figure 6).

PCANs developed for this study could be utilized for the purposeof delivering other morphogens (e.g., sonic hedgehog,12,49–52 bonemorphogenetic protein,53,54 and basic fibroblast growthfactor-812,49,55) to induce differentiation of hiPSCs into morespecific neuronal subtypes, such as motor, dopaminergic, GABAer-gic, and glutamatergic neurons. Combined delivery of these factorswith RA using PCANs would be expected to allow for morespecified neuronal lineage differentiation of hiPSCs.

hiPSC engineering combined with nanoparticle-based RAdelivery may have clinical implications for the treatment ofvarious neuronal disorders. hiPSCs can provide clinically-available cell sources for autologous cell therapy becausehiPSCs are prepared via epigenetic reprogramming of patient’ssomatic cells.56 In our study, neuronal differentiation ofhiPSC-derived neural rosettes was significantly promoted byintracellular RA delivery mediated by thermo-responsivePCANs (Figure 6). Therefore, PCAN-based RA delivery couldallow for autologous hiPSC therapy with enhanced therapeuticand regenerative efficacy for treating the patients with neuronaldisorders. In future work, in vivo therapeutic efficacy of hiPSCsengineered with PCAN-RA delivery system should be evaluatedin animal models of various neuronal disorders includingcerebral ischemia, neurodegenerative diseases, and traumaticneuronal injury. Given that thermo-responsive PCANs canrelease loaded cargoes at body temperature which is above LCST(35 °C) (Figure 1, E), PCANs may also be directly utilized for invivo delivery of varied growth factors and drugs, which allowsfor wider applications of PCANs for biomedical engineering.

In summary, we synthesized RA-loaded PCANs withenhanced cellular uptake ability. The deswelling of PCANswas increased with temperatures. This thermos-responsiveproperty facilitated RA release into the cells after cellular uptakein a physiologically relevant condition. In a neurogenesis modelusing hiPSCs, we demonstrated that hiPSCs cultured with1-2 μg/ml RA-loaded PCANs exhibited enhanced differentiationinto neuronal lineage cells. Therefore, this RA-loaded PCANcould be a powerful carrier for regulating neuronal differentia-tion of hiPSCs and treatment of neurodegenerative diseases.

References

1. Phanstiel DH, Brumbaugh J, Wenger CD, Tian S, Probasco MD, BaileyDJ, et al. Proteomic and phosphoproteomic comparison of human ESand iPS cells. Nat Methods 2011;8:821-7.

2. Bilic J, Izpisua Belmonte JC. Concise review: Induced pluripotent stemcells versus embryonic stem cells: close enough or yet too far apart?Stem Cells 2012;30:33-41.

3. Sanchez-Danes A, Consiglio A, Richaud Y, Rodriguez-Piza I, Dehay B,Edel M, et al. Efficient generation of A9 midbrain dopaminergic neurons

by lentiviral delivery of LMX1A in human embryonic stem cells andinduced pluripotent stem cells. Hum Gene Ther 2012;23:56-69.

4. Chung CY, Yang JT, Kuo YC. Polybutylcyanoacrylate nanoparticle-mediated neurotrophin-3 gene delivery for differentiating iPS cells intoneurons. Biomaterials 2013;34:5562-70.

5. Tang X, Zhou L, Wagner AM, Marchetto MC, Muotri AR, Gage FH, et al.Astroglial cells regulate the developmental timeline of human neuronsdifferentiated from induced pluripotent stem cells. Stem Cell Res2013;11:743-57.

6. Mohamad O, Yu SP, Chen D, Ogle M, Song M, Wei L. Efficientneuronal differentiation of mouse ES and iPS cells using a rotary cellculture protocol. Differentiation 2013;86:149-58.

7. Xu J, Wang H, Liang T, Cai X, Rao X, Huang Z, et al. Retinoic acidpromotes neural conversion of mouse embryonic stem cells in adherentmonoculture. Mol Biol Rep 2012;39:789-95.

8. Wilson L, Gale E, Chambers D, Maden M. Retinoic acid and the control ofdorsoventral patterning in the avian spinal cord. Dev Biol 2004;269:433-46.

9. Maden M. Retinoic acid in the development, regeneration andmaintenance of the nervous system. Nat Rev Neurosci 2007;8:755-65.

10. Santos T, Ferreira R, Maia J, Agasse F, Xapelli S, Cortes L, et al.Polymeric nanoparticles to control the differentiation of neural stem cellsin the subventricular zone of the brain. ACS Nano 2012;6:10463-74.

11. Akanuma H, Qin XY, Nagano R, Win-Shwe TT, Imanishi S, Zaha H, etal. Identification of stage-specific gene expression signatures in responseto retinoic acid during the neural differentiation of mouse embryonicstem cells. Front Genet 2012;3:141.

12. Krencik R, Weick JP, Liu Y, Zhang ZJ, Zhang SC. Specification oftransplantable astroglial subtypes from human pluripotent stem cells.Nat Biotechnol 2011;29:528-34.

13. Marchetto MC, Carromeu C, Acab A, Yu D, Yeo GW, Mu Y, et al. Amodel for neural development and treatment of Rett syndrome usinghuman induced pluripotent stem cells. Cell 2010;143:527-39.

14. Hirami Y, Osakada F, Takahashi K, Okita K, Yamanaka S, Ikeda H, etal. Generation of retinal cells frommouse and human induced pluripotentstem cells. Neurosci Lett 2009;458:126-31.

15. Martins AH, Resende RR, Majumder P, Faria M, Casarini DE, TarnokA, et al. Neuronal differentiation of P19 embryonal carcinoma cellsmodulates kinin B2 receptor gene expression and function. J Biol Chem2005;280:19576-86.

16. Sartore RC, Campos PB, Trujillo CA, Ramalho BL, Negraes PD,Paulsen BS, et al. Retinoic acid-treated pluripotent stem cells undergoingneurogenesis present increased aneuploidy and micronuclei formation.PLoS One 2011;6:e20667.

17. Szuts EZ, Harosi FI. Solubility of retinoids in water. Arch BiochemBiophys 1991;287:297-304.

18. Maia J, Santos T, Aday S, Agasse F, Cortes L, Malva JO, et al.Controlling the neuronal differentiation of stem cells by theintracellular delivery of retinoic acid-loaded nanoparticles. ACS Nano2011;5:97-106.

19. Ku BR, Kim JE, Chung BH, Chung BG. Retinoic acid-polyethyleneiminecomplex nanoparticles for embryonic stem cell-derived neuronal differen-tiation. Langmuir 2013;29:9857-62.

20. Park JS,YangHN,WooDG, JeonSY, ParkKH. Poly(N-isopropylacrylamide-co-acrylic acid) nanogels for tracing and delivering genes to humanmesenchymal stem cells. Biomaterials 2013;34:8819-34.

21. Li X, Zhou J, Liu Z, Chen J, Lu S, Sun H, et al. A PNIPAAm-basedthermosensitive hydrogel containing SWCNTs for stem cell transplan-tation in myocardial repair. Biomaterials 2014;35:5679-88.

22. Pereira CL, Goncalves RM, Peroglio M, Pattappa G, D'Este M, Eglin D,et al. The effect of hyaluronan-based delivery of stromal cell-derivedfactor-1 on the recruitment of MSCs in degenerating intervertebral discs.Biomaterials 2014;35:8144-53.

23. Tzouanas SN, Ekenseair AK, Kasper FK, Mikos AG. Mesenchymalstem cell and gelatin microparticle encapsulation in thermally andchemically gelling injectable hydrogels for tissue engineering. J BiomedMater Res A 2014;102:1222-30.

1869H.I. Seo et al / Nanomedicine: Nanotechnology, Biology, and Medicine 11 (2015) 1861–1869

24. Liao T, MoussallemMD, Kim J, Schlenoff JB, Ma T. N-isopropylacrylamide-based thermoresponsive polyelectrolyte multilayer films for humanmesenchymal stem cell expansion. Biotechnol Prog 2010;26:1705-13.

25. Gheisari Y, Yokoo T, Matsumoto K, Fukui A, Sugimoto N, Ohashi T, etal. A thermoreversible polymer mediates controlled release of glial cellline-derived neurotrophic factor to enhance kidney regeneration. ArtifOrgans 2010;34:642-7.

26. Tao ZM,Wang PZ,Wang L, Xiao L, Zhang FZ, Na J. Facile oxidation ofsuperaligned carbon nanotube films for primary cell culture and geneticengineering. J Mater Chem B 2014;2:471-6.

27. Soumen R, Johnston AH, Moin ST, Dudas J, Newman TA, Hausott B, etal. Activation of TrkB receptors by NGFbeta mimetic peptide conjugatedpolymersome nanoparticles. Nanomedicine 2012;8:271-4.

28. Tao Zhimin, Wang Peizhe, Zhang Fengzhi, Na J. Carbon nanotube thinfilm-supported fibroblast and pluripotent stem cell growth. Bio Nano Sci2014;4:288-300.

29. Wang C, Mallela J, Garapati US, Ravi S, Chinnasamy V, Girard Y, et al.A chitosan-modified graphene nanogel for noninvasive controlled drugrelease. Nanomedicine 2013;9:903-11.

30. PeiY, Chen J, YangL, Shi L, TaoQ, LiBHJ. The effect of pHon the LCSTof poly(N-isopropylacrylamide) and poly(N-isopropylacrylamide-co-acrylic acid). J Biomater Sci 2004;15:585-94.

31. Ono Y, Shikata T. Hydration and dynamic behavior of poly(N-isopropylacrylamide)s in aqueous solution: a sharp phase transition at thelower critical solution temperature. J Am Chem Soc 2006;128:10030-1.

32. Lebon F, Caggioni M, Bignotti F, Abbate S, Gangemi F, Longhi G, et al.Coil-to-globule transition of poly(N-isopropylacrylamide) doped withchiral amino acidic comonomers. J Phys Chem B 2007;111:2372-6.

33. Ringsdorf H, Sackmann E, Simon J,Winnik FM. Interactions of liposomesand hydrophobically-modified poly-(N-isopropylacrylamides): an attemptto model the cytoskeleton. Biochim Biophys Acta 1993;1153:335-44.

34. Graziano G. On the temperature-induced coil to globule transition ofpoly-N-isopropylacrylamide in dilute aqueous solutions. Int J BiolMacromol 2000;27:89-97.

35. Zhang XZ, Yang YY, Chung TS, Ma KX. Preparation and character-ization of fast response macroporous poly(N-isopropylacrylamide)hydrogels. Langmuir 2001;17:6094-9.

36. Zareiea HM, Bulmusb EV, Gunningc AP, Hoffmand AS, Piskinb E,Morrisc VJ. Investigation of a stimuli-responsive copolymer by atomicforce microscopy. Polymer 2000;41:6723-7.

37. Oliveira JM, Sousa RA, Malafaya PB, Silva SS, Kotobuki N, Hirose M,et al. In vivo study of dendronlike nanoparticles for stem cells “tune-up":from nano to tissues. Nanomedicine 2011;7:914-24.

38. Kim DS, Lee DR, Kim HS, Yoo JE, Jung SJ, Lim BY, et al. Highly pureand expandable PSA-NCAM-positive neural precursors from humanESC and iPSC-derived neural rosettes. PLoS One 2012;7:e39715.

39. Chambers SM, Fasano CA, Papapetrou EP, Tomishima M, Sadelain M,Studer L. Highly efficient neural conversion of human ES and iPS cellsby dual inhibition of SMAD signaling. Nat Biotechnol 2009;27:275-80.

40. Shi Y, Kirwan P, Livesey FJ. Directed differentiation of humanpluripotent stem cells to cerebral cortex neurons and neural networks.Nat Protoc 2012;7:1836-46.

41. Yang K, Jung H, Lee HR, Lee JS, Kim SR, Song K, et al. Multiscale,hierarchically patterned topography for directing human neural stemcells into functional neurons. ACS Nano 2014;8:7809-22.

42. Seo J, Lee JS, Lee K, Kim D, Yang K, Shin S, et al. Switchable water-adhesive, superhydrophobic palladium-layered silicon nanowires poten-tiate the angiogenic efficacy of human stem cell spheroids. Adv Mater2014;26:7043-50.

43. Park HJ, Yu SJ, Yang K, Jin Y, Cho AN, Kim J, et al. Paper-basedbioactive scaffolds for stem cell-mediated bone tissue engineering.Biomaterials 2014;35:9811-23.

44. Patil S, Sandberg A, Heckert E, Self W, Seal S. Protein adsorption andcellular uptake of cerium oxide nanoparticles as a function of zetapotential. Biomaterials 2007;28:4600-7.

45. Bonoiu AC, Mahajan SD, Ding H, Roy I, Yong KT, Kumar R, et al.Nanotechnology approach for drug addiction therapy: gene silencingusing delivery of gold nanorod-siRNA nanoplex in dopaminergicneurons. Proc Natl Acad Sci U S A 2009;106:5546-50.

46. Wilhelm C, Billotey C, Roger J, Pons JN, Bacri JC, Gazeau F.Intracellular uptake of anionic superparamagnetic nanoparticles asa function of their surface coating. Biomaterials 2003;24:1001-11.

47. Lim SJ, Kim CK. Formulation parameters determining the physico-chemical characteristics of solid lipid nanoparticles loaded with all-transretinoic acid. Int J Pharm 2002;243:135-46.

48. Castro GA, Orefice RL, Vilela JM, Andrade MS, Ferreira LA.Development of a new solid lipid nanoparticle formulation containingretinoic acid for topical treatment of acne. J Microencapsul2007;24:395-407.

49. Perrier AL, Tabar V, Barberi T, Rubio ME, Bruses J, Topf N, et al.Derivation of midbrain dopamine neurons from human embryonic stemcells. Proc Natl Acad Sci U S A 2004;101:12543-8.

50. Ye W, Shimamura K, Rubenstein JL, Hynes MA, Rosenthal A. FGF andShh signals control dopaminergic and serotonergic cell fate in theanterior neural plate. Cell 1998;93:755-66.

51. Hester ME, Murtha MJ, Song S, Rao M, Miranda CJ, Meyer K, et al.Rapid and efficient generation of functional motor neurons from humanpluripotent stem cells using gene delivered transcription factor codes.Mol Ther 2011;19:1905-12.

52. Xi J, Liu Y, Liu H, Chen H, Emborg ME, Zhang SC. Specification ofmidbrain dopamine neurons from primate pluripotent stem cells. StemCells 2012;30:1655-63.

53. Liu A, Niswande LA. Bone morphogenetic protein signalling andvertebrate nervous system development. Nat Rev Neurosci2005;6:945-54.

54. Ulloa F, Marti E. Wnt won the war: antagonistic role of Wnt over Shhcontrols dorso-ventral patterning of the vertebrate neural tube. Dev Dyn2010;239:69-76.

55. Streit A, Berliner AJ, Papanayotou C, Sirulnik A, Stern CD. Initiation ofneural induction by FGF signalling before gastrulation. Nature2000;406:74-8.

56. Robinton DA, Daley GQ. The promise of induced pluripotent stem cellsin research and therapy. Nature 2012;481:295-305.