Applications of electrospinning and supercritical carbon dioxide foaming techniques in controlled release and bone regeneration 7

CHAPTER 8 Lysine-Based Peptide Functionalized PLGA Foams for Controlled DNA Delivery† 8.1 Introduction Over recent years, DNA delivery research has become increasingly popular due to i

CHAPTER 8

Lysine-Based Peptide Functionalized PLGA Foams for

Controlled DNA Delivery†

8.1 Introduction

Over recent years, DNA delivery research has become increasingly popular due to its potential therapeutic and medicinal applications Among various delivery devices, microspheres and nanoparticles are the most widely used carriers in DNA delivery due to their uniform morphology and efficacy in cell transfection (Leong et al., 1998; MacLaughlin et al., 1998; Mao et al., 2001; Roy et al., 1997; Mao et al., 1996) There is also growing interest in the use of porous materials as DNA delivery matrices The stable and uniform porous structures, tunable pore size and well-defined surface properties of these materials allows the incorporation and release of a diversity of proteins and DNAs

in a more reproducible and predictable manner (Thomson et al., 1998; Chen et al., 2001;

structures can be easily molded into a desired shape, which can hold both DNA and cells simultaneously, and provide an appropriate platform for the formation and remodeling of new tissue with the degradation of the mold (Goldstein et al., 2001) Thus, porous foams appear to have significant advantages over microspheres and nanoparticles towards

† This chapter highlights the work published in H Nie, L.Y Lee, H Tong and C.H Wang Lysine-Based

Peptide Functionalized PLGA Foams for Controlled Gene Delivery J Control Release 2009

creating a DNA delivery and tissue engineering dual system Micro-porous PLGA foams engineered by supercritical carbon dioxide foaming technique with large, controlled pore size and highly ordered morphology offer an intriguing channel structure for DNA delivery and cell adhesion (Mikos et al., 1994) Furthermore, the sorption capacity and characteristics of micro-porous PLGA foam could be substantially altered by anchoring a variety of functional groups onto the external and internal pore surfaces Porous PLGA foam has been frequently used in drug and protein delivery (Hsu et al., 1996; Kim et al., 2006) However, its application in DNA delivery has been limited, mainly due to its negative surface charges, resulting in a strong charge repulsion that hinders the adsorption of DNA and attachment of normal cells onto the foams Therefore, surface functionalization of the PLGA foam is essential to convert it to an effective DNA carrier

to hold DNA and subsequently release it in a sustained manner PLGA/chitosan composite foams developed in Chapter 7 show promising results in controlled release of DNA, but the release rate of DNA and subsequent expression of target protein is too low, especially in the initial stage (Figures 7.5a and 7.7b) Therefore, an initial and significant release of DNA is demanded in order to optimize this kind of devices

Lysine, an α-amino acid of chemical formula HO2CCH(NH2)(CH2)4NH2, pKa 10.54 and hydrophobility of -3.9 (Civitelli et al., 1992), is a potentially good candidate as supplement for PLGA to fine-tune its charge property and hydrophilicity for DNA delivery purposes The primary amine side groups of lysine can interact and form complexes with DNA molecules In this study, the functionalization of PLGA porous foam matrix was accomplished using Lysine-based peptides It was speculated that the

functionalized foams may have different DNA loading and release profiles and thus cell transfection level, depending on the molecular properties of the peptides being used Particularly in this study, PLGA porous foams were functionalized using K4 and K20 peptides and the surface physical properties of the foams were investigated using a series

of state-of-the-art techniques, such as SEM and XPS BMP-2 plasmid was used as a model DNA and loaded onto foams with and without surface modification The

adsorption capacities of the foams and in vitro release of the model DNA in buffered saline (PBS) were studied In addition, cell proliferation on the foams and in vitro DNA expression were also investigated

phosphate-8.2 Materials and methods

8.2.1 Materials

Poly (D,L lactic-co-glycolic acid) (PLGA) containing a free carboxyl end group (uncapped)

with L/G molar ratio of 50:50 (PLGA 4A, MW=63k, IV=0.44) was purchased from Lakeshore Biomaterials (Cat W3066-603, AL, USA) Dichloromethane (DCM) (Cat No DR-0440) was purchased from Tedia Company Inc (Fairfield, OH, US.) Fmoc-Lys (Boc)-OH and phosphate-buffered saline (PBS) buffer containing 0.1 M sodium

phosphate and 0.15 M sodium chloride, pH 7.4., used for in-vitro study were purchased

from Sigma Aldrich (St Louis, MO, US) PreMix WST-1 cell proliferation assay system,

procured from Takara Bio Inc (Otsu, Shiga, Japan), Thermo Fisher Scientific Inc (Wilmington, DE, US) and R&D Systems (Minneapolis, MN, US), respectively

8.2.2 Preparation of foams and Lysine peptides

Blank PLGA foams were engineered based on a gas foaming method using supercritical

CO2 as the blowing agent All the procedures are the same as explained in Chapter 7 (see

Figure 7.1) Both peptides K-K-K-K-G (K4) and

K-K-K-K-K-K-K-K-K-K-K-K-K-K-K-K-K-K-K-K-G (K20) (where K and G represents Lysine and glycine residue, respectively) were synthesized in-house on an automated Multipep peptide synthesizer (Intavis, Germany) All peptides were assembled on Fmoc-Glycine resin (substitution level = 0.66 mmole/g resin) at 50 μmole scale Stepwise couplings of amino acids were accomplished using double coupling method with 5-fold excesses of amino acids, equivalent activator reagents, 2-(1H-Benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate and N-Hydroxybenzotriazole, and two equivalents of base, N-methylmorpholine The removal of Fmoc was completed using 20% piperidine in dimethylformamide (DMF) Cycles of deprotection, washings, double couplings, and washings were repeated until the desired chain length was achieved The dried peptidyl-resin was cleaved by a cocktail solution composed of 95% trifluoroacetic acid (TFA), 2.5% deionized water, and 2.5% triethylsilane (v/v) The crude peptide was purified using an Agilent 1100 semi-preparative high performance liquid chromatography (HPLC) (Santa Clara, CA) The purification was performed on an Agilent Zorbax 300SB-C18 reverse phase (RP) column (5 μm particle size, 300Ǻ pore size, 25 x 1.0 cm) with a linear gradient of buffer A (0.1% TFA in water) and buffer B (0.1% TFA in acentonitrile) from 10% B to 45% B in 30 min

at a flow rate of 4 mL/min The purity of all peptides was greater than 95% by analytical RP-HPLC and matrix-assisted laser desorption/ionization-time of flight mass

spectroscopy (MALDI-TOF MS) on a Bruker AutoFlex II MALDI-TOF MS (Bruker, Bremen, Germany) (data not shown)

8.2.3 Peptides conjugation

K4 and K20 were employed to study the effects of chain length and surface charges on the adsorption and release patterns of plasmid DNA Blank PLGA foams were sterilized with 70% ethanol and washed thrice with excess sterilized water To functionalize the PLGA foams, the peptides (K4 or K20) were incorporated covalently onto the surface of the PLGA foams using a condensation coupling method (Li et al., 1998) Briefly, the

carboxyl groups on the foam surface were first activated by 10mM of

N-Ethyl-N'-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC hydrochloride, Sigma-Aldrich)/

N-Hydroxysulfosuccinimide (NHS, Aldrich) sterilized by filtering through 0.22 μm filter

for 5 h with occasional shaking at room temperature The foams were then washed 3 times with sterilized water to eliminate excess EDC/NHS The peptides (K4 or K20) were covalently immobilized onto the activated foams by immersing the foams in the peptide solutions (0.1mM) at room temperature overnight (Khew et al., 2007) After that, the unbound K4 or K20 was desorbed in copious amounts of PBS for 1 h at room temperature The resultant foams were thoroughly washed with DI water and dried in air

8.2.4 Characterization of morphology and porosity

The morphology of samples (blank foams, F0; K4-functionalized foams, F1; functionalized foams, F2) was examined using scanning electron microscopy (SEM) (JSM 5600LV, JEOL) The porosity of the porous pure PLGA foam and the modified

K20-PLGA foams was measured using a mercury intrusion porosimeter AutoPore III 9420 (Micromeritics, Norcross, GA) (Zhang et al., 2009)

8.2.5 Atomic composition of foam surface

The chemical structure and atomic composition of the blank and surface-modified foams were characterized using X-ray photoelectron spectroscopy (XPS) (VG ESCALAB 220I-XL; Thermo VG Scientific, UK), with the data processing performed using XPSPEAK (Version 4.1) software Wide scan (0-1000 eV) and high-resolution (C1s, O1s, and N1s) spectra were acquired, respectively

8.2.6 Plasmid preparation and loading procedures

A pT7T3D-PacI encoding BMP-2 was used in this study The plasmid DNA was

amplified in a transformant of Escherichia coli bacteria and isolated from the bacteria by

Corporation, MD, USA) The DNA concentration was determined using a Thermo

plasmid DNA, different kinds of foams (F0, F1 and F2) were introduced into 0.5 mL of

TE buffered solution of DNA (200 μg/mL) and soaked for 24 h under constant stirring The foams were then dried under vacuum after quick and through washes by DI water

8.2.7 DNA adsorption capacity on foams

Besides the atomic composition analysis, the densities of DNA attracted on different foams were quantified Briefly, 15 mg of each scaffold was dissolved in 0.5 mL of DCM

and 2 mL of PBS (pH 5.0) was introduced to scaffold/DCM solution, vortexed, and centrifuged (Hettich Zentrifugen, Universal 32R, Andreas Hettich GmbH & Co KG, Tuttlingen, Germany) at 14,000 rpm for 3 min The aqueous layer was collected, and two more extraction cycles were performed to maximize DNA recovery The water phases were kept frozen at -20 °C until they were analyzed for DNA concentrations using Thermo Scientific NanoDropTM 1000 Spectrophotometer

8.2.8 In vitro DNA release studies

Foams were sterilized using Co-60 gamma irradiation at a dose of 15 kGy before using

for DNA adsorption, and following DNA in vitro release and cell culture studies The in vitro release of plasmid DNA was carried out over a period of 20 days and the cumulative

release curve was plotted Foams of 5 mg each loaded with plasmid was added to 1 mL of PBS (pH=7.4) and the resultant solution was then placed in an orbital shaker bath (GFL®

solution at specific intervals and then topped up with 0.1 mL of fresh media Each study group (F0, F1 and F2) was tested in triplicate and all the collected samples were stored at -20 °C until the release assay The DNA concentration in each sample was determined by Thermo Scientific NanoDropTM 1000 Spectrophotometer

To evaluate the effects of charge interaction on the molecular integrity of plasmid DNA, agarose DNA gel electrophoresis was used to determine the integrity of plasmid DNA

released from the foams in vitro after 5 days Release samples were diluted six-fold in

Blue/Orange Loading Dye (Promega, Madison, WI, US) A 12 μL of loading

buffer/sample was loaded into each well of 1% agarose gel Electrophoresis was conducted using a Bio-Rad Mini-PROTEAN III electrophoresis system (Bio-Rad Laboratories, CA, US) at a constant voltage (60V) for 120 min with native plasmid DNA

as control SYBR Gold staining (Molecular Probes, Invitrogen) was employed to stain plasmid in samples/control and Gene Genius Bio Imaging system (Syngene, Cambridge, UK) was used to image the gels

8.2.9 Preparation and culture of rat marrow stromal cells

The seeds of rat marrow stromal cells (rMSCs) used in the current study were donated from the orthopaedic research center, Kaohsiung Medical University as a gift They were cultured in DMEM supplemented with 4mM-glutamine (Biological Industries, Kibbutz Beit Haemek, Israel), 25 mM HEPES buffer, 10% fetal bovine serum (Gibco), 10U/mL penicillin G sodium and 10 mg/mL streptomycin as Fungizone (Gibco) and incubated at

extracted with PBS solution containing 0.25% trypsin-EDTA (Biological Industries, Kibbutz Beit Haemek, Israel) and normally subcultured at a density of 2 x 104 cells/cm2

8.2.10 Cell viability assay

100 μL of rMSCs suspension (1 x 105 cells/mL) along with different foams were added into wells of 96-well plates (NunclonTM, Roskilde, Denmark) and incubated at 37 °C and

same conditions (without foam) was denoted as a control At specific intervals (on the first, second and third day), cell viability was measured using a standard cell proliferation

assay (PreMix WST-1 cell proliferation assay system, Takara Bio Inc, Shiga, Japan) The cell viability was calculated as following (Takashima et al., 2007):

Cell viability (%) = (Abs test cells/Abs control cells) x 100% (8.1)

Where “Abs test cells” represents the amount of formazan produced by cells treated with the different formulations and “Abs control cells” represents the amount of formazan produced by cells in the control

8.2.11 In vitro experiment of cells transfection

100 μL of rMSC suspension (1 x 105 cells/mL) was added into wells of 96-well plates (NunclonTM, Roskilde, Denmark) and incubated for 16 h for adherence Afterwards, the media was aspirated from the wells and the wells were washed once with DMEM before

100 μL of new DMEM was added to each well along with different foams as described in the cytotoxicity experiment To measure the level of gene transfection of rMSC cultured, the cells were washed three times with PBS, and homogenized in the lysis buffer (0.1M Tris-HCl, 2mM EDTA, 0.1% Triton X-100) After staying in ice for 10 mins, the sample lysate (100 μL) was centrifuged at 12,000 rpm for 5 min at 4 °C, and the supernatant was carefully collected and kept in the ice To measure the expression level of BMP-2 gene,

50 µl of the supernatant was collected and the 2 protein was determined by a

BMP-2 ELISA Kit (R&D Systems, US) All transfection experiments were performed at determined intervals and assayed in triplicate (Nie and Wang, 2007; Li et al., 2003)

pre-8.2.12 Statistic analysis

All data are presented as mean ± S.D throughout this study Statistical analysis of the experimental data was performed and α < 0.05 is considered as significantly different

Figure 8.1 SEM images of blank foam (F0) and functionalized foams (F1 and F2)

8.3 Results and discussion

8.3.1 Surface characterization and porosity measurement of foams

As a dual system for tissue engineering and DNA delivery, the porous structure can provide both sufficient space for blood circulation and also extended surface area for the entrapment of large amount of DNA Figure 8.1 shows typical SEM morphologies of F0, F1 and F2 and the 3D inter-connected porous structures are evident Three foams from the same batch were measured and the average value (with a sampling size of 100 pores) was used to indicate the diameter From the SEM images of F0, the pore diameters of the foams are relatively uniform and they all fall within the range of 20.8-59.5 µm After conjugation of peptides, the pore shapes become irregular and the inter-connected porous structures are modified as well Some pores are isolated and not connected to other pores

in F0 In contrast, all pores in F1 and F2 are open and interconnected The changes in pore structures are ascribed to the activation of the carboxyl groups by EDC/NHS, as similar changes in pore structures are also observed in foams treated by EDC/NHS alone, prior to the incorporation of K4/K20 Actually the interconnectivity of blank foams is not

so good and many pores are blocked by thin membranes, as shown by the arrows in Figure 8.1 However, the membranes are very weak and easy to be damaged by the harsh environment imposed by EDC/NHS The destruction of pores modifies the structures and increases the interconnectivity of foams Table 8.1 shows the initial porosity of F0 and also the foams after going through the surface modifications by K4 or K20 As an evidence of structural changes after the conjugations of peptides, the porosities of F1 and F2 are slightly higher than F0 This result confirmed that the process of lysine

modifications on foams did slightly change the interconnectivity of pores and create more channels in the three-dimensional structure

8.3.2 XPS spectra of modified surfaces

Figure 8.2a illustrates the C1s high-resolution XPS spectra of the foams before DNA loading The C1s spectrum of F0 shows the characteristic peak of C–C/C–H, C-O and C=O bonds with binding energies of 284.8 eV, 287 eV and 289.1 eV respectively In contrast, the spectrum of F1 conjugated with K4 (Figure 8.2b) showed that the two C1s peaks at around 284.8 eV and 287 eV were perturbed by other peaks After peak-deconvolution, one peak corresponding to C-N centered at 286.4 eV was observed When the conjugation peptide was changed from K4 to K20, significant increase of C-N peak was detected (Figure 8.2c) The presence of C-N peak displays the successful conjugation

of peptides on both F1 and F2 Furthermore, the successful grafting of peptides on foam surface was also verified by the presence of nitrogen (N1s) peaks at 397.9 eV from the N1s high-resolution XPS spectra

As shown in Figure 8.3a, significantly higher peaks of N1s were detected in F1 and F2 than that in F0 The N1s signal (before DNA adsorption) is directly associated with lysine peptides, so the percentages of C-N peak areas were well correlated with the nitrogen atomic concentration as indicated in Table 8.2 Similarly, the P2p signal is directly corresponding to the DNA on foams (after DNA adsorption), so an additional element P was detected on all foams after DNA loading process as shown in Figure 8.3b The atomic percentage of P in F0, F1 and F2 are 0.48, 2.23 and 1.89 respectively, which are