Microfluidic Synthesis and Angiogenic Activity of ... - ACS Publications

Subscriber access provided by Northern Illinois University

Article

Microfluidic Synthesis and Angiogenic Activity of Ginsenoside Rg1-loaded PPF Microspheres Mehrnaz Salarian, Raziye Samimi, William Z. Xu, Zhiqiang Wang, Tsun-Kong Sham, Edmund M. K. Lui, and Paul A. Charpentier ACS Biomater. Sci. Eng., Just Accepted Manuscript • DOI: 10.1021/acsbiomaterials.6b00222 • Publication Date (Web): 17 Aug 2016 Downloaded from http://pubs.acs.org on August 22, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Biomaterials Science & Engineering is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Biomaterials Science & Engineering

Microfluidic Synthesis and Angiogenic Activity of Ginsenoside Rg1-loaded PPF Microspheres

Mehrnaz Salarian1,2, Raziye Samimi2,3, William Z. Xu3, Zhiqiang Wang4, Tsun-Kong Sham4,5, Edmund M. K. Lui2,6, and Paul A. Charpentier*2, 3

1

Biomedical Engineering Graduate Program, University of Western Ontario, London, Ontario

2

The Ontario Ginseng Innovation & Research Consortium, London, Ontario

3

Chemical and Biochemical Engineering Department, University of Western Ontario, 1151 Richmond Street, London, Ontario, Canada N6A 5B9 4

Department of Chemistry, University of Western Ontario, London, Ontario

5

Soochow-Western Centre for Synchrotron Radiation Research, University of Western Ontario, London, Ontario 6

Department of Physiology and Pharmacology, University of Western Ontario, London, Ontario

*Author to whom correspondence should be addressed. E-mail: [email protected]; Phone: (519) 661-3466; Fax: (519) 661-3498

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 37

ABSTRACT Next generation polymer scaffolds for hard tissue engineering require unique structures to both enhance release kinetics while enabling bone cell growth (osteogenesis). This study examined the encapsulation of the pro-angiogenic mediator, ginsenoside Rg1, into biodegradable poly(propylene fumarate) (PPF) microspheres to facilitate osteogenesis, while examining the release mechanism using advanced X-ray absorption near edge structure spectroscopy (XANES). Ginsenoside Rg1-loaded PPF microspheres were prepared using both an emulsion method and a microfluidic device, with the microfluidic technique providing tunable unimodal PPF spheres ranging in size from 3-52 µm by varying the flow rates. The morphology and composition of the Rg1-loaded PPF microspheres were characterized using FTIR, XRD and XANES to examine the distribution of ginsenoside Rg1 throughout the polymer matrix. Encapsulation efficiency and release profiles were studied and quantified by UV-Vis spectrophotometry, showing high encapsulation efficiencies of 95.4 ± 0.8% from the microfluidic approach. Kinetic analysis showed that Rg1 release from the more monodisperse PPF microspheres was slower with a significantly smaller burst effect than from the polydisperse spheres, with the release following Fickian diffusion. The released Rg1 maintained its angiogenic effect in vitro, showing that the PPF microspheres are promising to serve as vehicles for long-term controlled drug delivery leading to therapeutic angiogenesis in bone tissue engineering strategies.

KEYWORDS: Poly(propylene fumarate); Monodisperse microspheres; Angiogenesis; Drug delivery; Microfluidics; X-ray absorption near edge structure (XANES).


2

Page 3 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

INTRODUCTION For bone tissue engineering, new approaches are required to produce polymer scaffolds that

not only have sufficient mechanical properties but are also biocompatible and biodegradable.1 In addition, these matrices must support cellular migration and proliferation to allow for the infiltration of osteoblasts and the formation of bone.2-3 Current commercial systems are based on acrylic cements such as poly(methyl methacrylate) (PMMA) and calcium phosphate.4 Several disadvantages are associated with these formulations including having no biodegradability or bioactivity, while providing no angiogenesis activity for bone cell growth.5 A promising polymeric scaffold matrix that can potentially provide sufficient mechanical properties while overcoming these disadvantages is poly(propylene fumarate) (PPF). PPF is an unsaturated linear polyester that can be cross-linked through its carbon-carbon double bonds, and degraded by simple hydrolysis of the ester bonds into nontoxic products (Figure 1a).6-8

PPF can also be

synthetically modified to covalently bind to hydroxyapatite, minerals or potentially angiogenic agents through its carbon-carbon double bonds, improving its mechanical and release properties.9 Another advantage of PPF over many other biodegradable synthetic polymers is that it can be utilized as an injectable system, allowing for direct injection into a defect site and cross-linking in situ.10 Commercial acrylic systems require controlled injectability and resin flow properties during surgery.11 Polymeric microspheres of controlled dimensions have the potential to enhance the injectability of the microsphere/scaffold composite12 although are relatively unexplored for bone tissue engineering. This enhanced injectability can help control local delivery of the active agent, enhancing the therapeutic response while reducing the incidents of peak-related side effects and systemic toxic effects.13 Furthermore, tuning the size of the microspheres can control their release kinetics over a prolonged period of time; preventing the requirement for multiple doses


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 37

during prolonged therapy.13 For forming these microspheres, microfluidics has emerged as a flexible approach to control the sphere size14-15 while producing microspheres with high-drug loading and encapsulation efficiency, and sustained release behavior due to outstanding monodispersity.16 In addition to providing for the required flow rheology during surgery/treatment, enhancing microvascularization and microcirculation is critical for enhancing bone regeneration.17 For forming the required new blood vessels, angiogenesis is a complex multi-step process involving capillary tube formation of endothelial cells.18 The influence of angiogenesis on osteogenesis has been previously demonstrated, with the vasculature providing enhanced mass transport for the tissue, and delivering circulating stem cells to participate in bone formation.19 Among the angiogenic regulators, vascular endothelial growth factor (VEGF) and nitric oxide (NO) work closely with one another in the modulation of angiogenesis. Interestingly, ginsenoside Rg1 (Figure 1b), which can be isolated from ginseng root, has been shown to be a highly stable proangiogenic compound which stimulates angiogenesis through enhancing the production of NO and up-regulating the VEGF expression.20-21 According to mechanistic studies, the responses were mediated through the PI3K → Akt pathway.22 Angiogenesis induced by ginsenoside Rg1 was found to be comparable to or even better than that induced by bFGF in vivo. Rg1 was also found to promote functional neovascularization into a polymer scaffold in vivo.23 We hypothesized that a synthetic system based on PPF microspheres could control ginsenoside Rg1 loading and release while providing regeneration of vascular tissue in a localized and sustained manner. To the best of our knowledge, no previous study has addressed the fabrication of PPF microspheres via the microfluidic method. In addition, very little is known about the interaction between the Rg1 molecules and the PPF-based carriers. For the first time this paper


4

Page 5 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


reports the use of C K-edge XANES spectroscopy, which is a type of synchrotron radiation techniques, to identify how ginsenoside Rg1 molecules interact with PPF microspheres. The hypothesis underlying this study is that delivery of the pro-angiogenic mediator ginsenoside Rg1 from PPF microspheres can provide sufficient angiogenesis to enable bone regeneration in osseous defects. The following two questions are addressed in the present study: (1) how does the microsphere preparation process influence the size distribution and encapsulation efficiency of Rg1 in the prepared PPF microspheres, as well as the kinetics and mechanism of release of Rg1? (2) how does the pro-angiogenic mediator, ginsenoside Rg1, maintain its angiogenic effect before and after encapsulation into the PPF microspheres?

a

HO

b

OH O OH

O HO

H

O

O O

OH

H

OH OH

H

H

HO

n

O

O

HO

OH OH

OH

Figure 1. Chemical structures of (a) PPF and (b) ginsenoside Rg1.


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

2

Page 6 of 37

EXPERIMENTAL SECTION

2.1 Materials 1,2-propanediol

(≥99.5%), diethyl fumarate (98%), zinc chloride anhydrous (ZnCl2,

≥99.995%), hydroquinone (99%), hydrogen chloride hydrochloric acid (37%), anhydrous magnesium sulfate (99.0%), anhydrous ethyl ether (≥99.0%), anhydrous methylene chloride (≥99.8%), poly(vinyl alcohol) (87–89% hydrolyzed, Mw=13,000–23,000 g/mol), HPLC grade methanol (≥99.9%), and 0.1 mol/L phosphate-buffered saline solution (PBS, pH 7.4) were purchased from Sigma-Aldrich Canada. Ginsenoside Rg1 was donated by Dr. Edmund Lui, Department

of

Physiology

and

Pharmacology,

University

of

Western

Ontario.

Poly(dimethylsiloxane) (PDMS; Sylgard 184) was supplied by Dow Corning (Midland, USA). Double distilled water used throughout the experiments was generated from a Milli-Q water purification machine (18.2 MΩ·cm resistivity, Barnstead EasyPureII, Thermo Scientific, USA). Glass syringes (1000 series, Hamilton Co., NV) and syringe pumps (KD scientific pump, Inc. Legato 2200, 1100) were used for sending liquids into microchannels. Human umbilical vein cell lines (EA.hy926) were provided by Dr. Jeff Dixon, Department of Physiology and Pharmacology, University of Western Ontario. Growth factor-reduced Matrigel was purchased from VWR. Cell culture medium and reagents were purchased from Gibco laboratories.

2.2 Methods 2.2.1

Synthesis of poly(propylene fumarate) (PPF)

PPF was synthesized using a two-step transesterification/polymerization method as described in Salarian et al.7 1,2-propanediol and diethyl fumarate were reacted under nitrogen where ZnCl2 and hydroquinone were used as a catalyst and a crosslinking inhibitor, respectively. In the first phase of the reaction, the temperature was gradually increased from 110 to 150 ºC in increments


6

Page 7 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


of 10 ºC every 15 min. Bis(hydroxypropyl) fumarate intermediate and ethanol, which was collected as a distillate were obtained. The reaction was stopped after ~90% of the theoretical yield of ethanol was collected. The second phase of the reaction was conducted under reduced pressure with a gradual increase in temperature from 100 to 150 ºC in an increment of 10 ºC every 30 min forming propylene glycol as a byproduct. The second phase was continued until PPF of a proper molecular weight was produced. Then, the resulting polymer was purified by several washes with HCl solution, distilled water, and brine, respectively. Magnesium sulfate was then employed for drying the organic polymer phase. After that, the polymer was concentrated by rotoevaporation and then precipitated in diethyl ether. The diethyl ether was decanted and the purified polymer was vacuum dried to remove any residual solvent.

2.2.2

Preparation of PPF microspheres encapsulating ginsenoside Rg1 using W/O/W

double emulsion method A water-in-oil-in-water (W/O/W) double emulsion-solvent extraction technique24, which includes encapsulation of water-soluble therapeutic agents for targetable drug delivery,25 was employed to prepare PPF microspheres encapsulating ginsenoside Rg1, as illustrated in Figure 2. Briefly, 100 mg of PPF was dissolved in 10 mL of dichloromethane, and 10 mg of ginsenoside Rg1 was dissolved in 5 mL of solution (methanol/double distilled water = 4:6 v/v). The ginsenoside Rg1 solution was added to the PPF solution to create the first emulsion W/O. After stirring for 2 h using a magnetic stirring bar at 1200 rpm, the whole mixture was re-emulsified in 20 mL of an aqueous solution of PVA (4% w/v) producing W/O/W double emulsion. After stirring for 0.5 h using a magnetic stirrer bar at 400 rpm, the emulsion was poured into a beaker containing 40 mL of aqueous PVA solution (0.5% w/v) and stirred for 2 h. This solution was then added to double-distilled water and stirred for another 3 h at room temperature.


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 37

Subsequently, the emulsion was ultrasonically agitated for 10 min at 30 °C. After removing DMC by evaporation, the formed microspheres were centrifuged at 5000 rpm for 5 min, followed by washing with distilled water twice to remove PVA and freeze-drying to remove water. The Rg1-loaded microspheres were stored at −20°C.

Figure 2. Schematic diagram of the double emulsion method.

2.2.3

Preparation of PPF microspheres encapsulating ginsenoside Rg1 using droplet

microfluidic method Microfluidic chips with typical channel dimensions of 100 µm in depth and 200 µm in width were fabricated using conventional soft lithography techniques as previously described.26 The schematic drawing of the microfluidic device is shown in Figure 3. The disperse phase was prepared by emulsifying the ginsenoside Rg1 solution [2 mg/mL or 10 mg of Rg1 in 5 mL of solution of methanol/double distilled water (2:3 v/v)] in PPF solution (100 mg of PPF in 10 mL of DCM), which flowed into inlet 1. The continuous phase, an aqueous solution of PVA (4% w/v), flowed into inlet 2. Both liquid phases were loaded into glass syringes, and digitally controlled syringe pumps delivered to the microfluidic device using Teflon tubing (1.6 mm OD,


8

Page 9 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


0.6 mm ID). Following the previous studies in our group, the flow rate of the continuous phase ranged between 0.05-0.2 mL/min while the flow rate of the dispersed phase was constant and held at 0.01 mL/min.26 The outlet of the device was a Teflon tubing submerged into a 10 mL beaker containing 0.5% w/v of PVA aqueous solution to collect the monodisperse droplets. Next, the solvent within the droplets was removed by evaporation at room temperature, and hardened microparticles were then centrifuged at 5000 rpm for 5 min, washed twice with distilled water to remove excess PVA, freeze-dried, and stored at −20 ºC.

Figure 3. Schematic diagram of the microfluidic system. (1) disperse phase inlet (ginsenoside Rg1 solution emulsified in PPF in DCM solution), (2) continuous phase inlet (4% w/v PVA aqueous solution), and (3) outlet.

2.3 Characterization 2.3.1

Characterization of PPF microspheres encapsulating ginsenoside Rg1

The surface morphology of microspheres was analyzed using scanning electron microscopy (SEM) (Hitachi S-4500) with samples prepared by applying the produced powders directly to a carbon adhesive tape. Samples were sputter-coated with gold to reduce charging problems. The particle size distribution was characterized by measuring the particle diameter in the SEM images using Image J software (version 1.37, National Institute of Health, USA), with the mean


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 37

diameters and standard deviations of over 100 microspheres being reported. ATR-FTIR (Nicolet 6700, Thermo Scientific) was used to investigate the chemical structure of the PPF, Rg1, and Rg1-loaded PPF microspheres. The spectra were recorded in the range of 600-4000 cm-1 with a resolution of 4 cm-1 over 32 scans. To evaluate the crystallinity of Rg1 after encapsulation, XRD data for Rg1, placebo PPF microspheres, and Rg1-loaded PPF microspheres were obtained using an Inel CPS Powder Diffractometer. The instrument was equipped with a Cu-X-ray radiation Tube, an Inel XRG3000 generator, and an Inel CPS 120 detector. The CPS is a curved detector and performed the simultaneous collection of X-rays diffracted by the samples over 120 ° (2θ). Powder samples were ground to a uniform composition and placed on a flat aluminum sample holder. To investigate the interaction between the Rg1 and PPF, C K-edge X-ray absorption near edge structures (XANES) of the Rg1 molecules, PPF microspheres, and Rg1-loaded PPF microspheres were measured at the high-resolution spherical grating monochromator (SGM) beamline at the Canadian Light Source (CLS), the University of Saskatchewan. The samples were mounted on indium foils with an angle of 20° facing toward the incident photon beam. XANES spectra were recorded in partial fluorescence yield (PFY) that was measured by detecting the X-ray fluorescence photons emitted from the element of interest (e.g.: carbon). All XANES spectra were normalized to the incident photon flux (I0), which was measured using a silicon wafer as the photodiode. To determine encapsulation efficiency (EE) for Rg1-loaded PPF microspheres prepared using the conventional double emulsion and microfluidic methods, 5 mg dried Rg1-loaded microspheres were dissolved in 1 mL of 1N NaOH solution.13 The Rg1 concentration was determined by measuring the absorption of the solution at the λmax value of 189 nm using a Shimadzu UV-Vis 3600 spectrophotometer. A calibration curve was obtained from Rg1 solutions


10

Page 11 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


with known concentrations (see supporting information Figure S1). Encapsulation efficiency of Rg1 in PPF microspheres was calculated by normalizing the quantity of Rg1 actually entrapped to the starting quantity of Rg1 ,12 i.e., the ratio of the actual Rg1 content of the microspheres to the amount initially added to them.27 All experiments were repeated three times. The EE of drug is defined as: %

100

(1)

with Wrec being the weight of Rg1 recovered from the PPF microspheres as measured by UV spectroscopy and Wadd is the weight of Rg1 added initially to the PPF microspheres.

2.3.2

In vitro release of ginsenoside Rg1

For the release study, 5 mg of Rg1-loaded PPF microspheres were dispersed into 10 mL of pH 7.4 PBS in a centrifuge tube. The tube was sealed and put in a shaking water bath with a rotation rate of 100 rpm at 37 °C. At specific time spans, the tube was centrifuged at 2500 rpm for 5 min, and 2 mL of the supernatant were collected and substituted by fresh PBS. The collected supernatants were stored at -20 °C. The absorption spectra at 189 nm wavelength were calculated with a UV-Vis spectrophotometer (Shimadzu UV-Vis 3600) to determine the amount of released Rg1. Cumulative release of Rg1 profile was acquired with the volume loss correction and plotted against time. All the experiments were repeated in triplicate for each time interval. The collected extracts were also used for the following bioassay studies.

2.3.3

Angiogenesis activity of ginsenoside Rg1 before encapsulation into and after

release from the PPF microspheres Tube formation bioassays were performed using the EA.hy926 cell line derived from the fusion of human umbilical vein endothelial cells (HUVEC) with the A549 human pulmonary adenocarcinoma cell line. The HUVE cell line has shown to attach, migrate, and form tubular structure after seeding on a Matrigel substrate.28-29 Cell lines were cultured in Dulbecco’s


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 37

modified Eagle medium (DMEM) supplemented with ECGS (20 mg/mL), 10% fetal buvine serum (FBS), 1% antibiotic, and 1 mL (5X) per 50 mL of medium Hat media supplement. The cells were grown at 37 °C in humidified air with 5% CO2 incubator. When the cell cultures were 80%-90% confluent, cells were harvested following trypsin treatment, and the released cells were resuspended to the cell density required for tube formation assay. To examine the angiogenic activity of ginsenoside Rg1 in vitro, 4.5×104 cells/well of EA.hy926 cell line were seeded in a growth factor-reduced Matrigel-coated 96-well tissue culture plate loaded with medium. Various concentrations of ginsenoside Rg1 were examined before encapsulation into the PPF microspheres and after release from the microparticles. After incubation for eighteen hours, images from a total of five microscopic fields per well were captured by an inverted microscope (Nikon TMS, Japan) using a 40× objective. Motic Image Plus 2.0 software (Motic Instruments Inc., Richmond, BC, Canada) was used for the analyzing the formation of tubes. The angiogenic effect was estimated by counting the branch points of the formed tubes, and then the average numbers of branch points were calculated. Each experiment was repeated at least three times, with each experiment yielding essentially identical results. Additional quantitative analysis of tube formation about tube length, number of tubes, number of nodes, number of loops, and loop area was performed on a single image of each sample using Wimasis Image Analysis (Ibidi GmbH, Germany).

2.3.4

Model fitting of in vitro release of Rg1

To investigate the kinetics of Rg1 release from the PPF microspheres, four commonly used empirical models were used to fit the in vitro release data, i.e., the Hixson−Crowell, Korsemeyer−Peppas, Weibull, and first-order models.30-32 The curve fitting toolbox of Matlab R2015b software was used to determine both the regression coefficients, r2, and the best fit model parameters to understand the release mechanism for the studied systems.


12

Page 13 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3

RESULTS AND DISCUSSION

3.1 Surface morphology and encapsulation efficiency of the Rg1loaded PPF microspheres SEM images of the PPF microspheres encapsulating ginsenoside Rg1 prepared using both a conventional W/O/W double emulsion method and a microfluidic approach are displayed in Figure 4. The sizes of individual microspheres were measured manually by means of Image J software with the average size and size distribution determined. The results in Figure 4a show that polydisperse microspheres of 2-45 µm with an average diameter of 13.5 µm and a standard deviation of 11.1 µm were obtained using the conventional double emulsion method. Figure 4b shows SEM images of PPF microcapsules prepared using the microfluidic method, when the flow rate for the continuous and disperse phases were set at 0.1 and 0.01 mL/min, respectively. Very uniform spherical and discrete microspheres (50-65 µm) with an average diameter of 52.0 µm and a standard deviation of 7.0 µm were formed by using the microfluidic device. In addition, the mean diameter of the microparticles was easily regulated by varying the flow rates of the continuous and disperse phases. For example, with a flow rate of 0.2 and 0.01 mL/min for continuous and disperse phases, respectively, narrowly distributed microspheres were formed with an average diameter of 3.0 µm and standard deviation of 1.8 µm (Figure 4c). However, lowering the flow rate of the continuous phase to 0.05 mL/min, the system could not produce any droplets and microspheres. In explanation, the droplet size at the T-junction is generally determined by the interfacial tension between the dispersed and the continuous phases, the superficial viscosity and velocity of the continuous phase.33 When the interfacial tension and superficial viscosity are fixed such as in this study, the droplet size can be tuned by varying the velocity of the continuous phase. Here, increasing the flow rate of the continuous phase would


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 37

lead to increased shear forces in comparison to surface tension forces, resulting in the observed smaller droplet formation.33 As PPF is a polyester with a significant amount of ester bonds (Fig. 1a), interaction of this semi-hydrophobic polymer with the aqueous phase via hydrogen bonding is possible, helping PPF act as a surfactant by forming a semi-liquid continuous film at the interface34 while the droplets size was reduces by high-speed stirring. If the flow rate of the continuous phase is too low, such as 0.05 mL/min in the present study, the droplets may elongate in the main channel with the developing neck starting to thin, failing to form spherical particles.16 The size of these obtained microspheres is ideal for mixing the prefabricated components in a syringe and fabricating microsphere/scaffold composites, as microspheres with diameter ranging from 1 to 110 µm are considered to be excellent, and the resulting composite could be effortlessly injected all the way through a small needle (14-gauge needle)12, while commercial formulations use a sieving approach to create monodisperse particles in the 50-80 µm size range.35


14

Page 15 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4. SEM images (left) and corresponding statistical analyses (right) of Rg1-loaded PPF microspheres from (a) double emulsion method and (b) microfluidic method where the flow rates of continuous and disperse phases were 0.1 and 0.01 mL/min, respectively, and (c) microfluidic method where the flow rates of continuous and disperse phases were 0.2 and 0.01 mL/min, respectively.


15


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 37

Encapsulation efficiency (EE) is an important factor in drug delivery, especially for valuable and expensive bioactive compounds.16 The EE of Rg1 was measured for PPF microspheres fabricated using both the microfluidic technique and double emulsion method and is summarized in Table 1. Rg1-loaded PPF microspheres prepared by the double emulsion method had an EE value of 78.5%. In contrast, the EE value of the PPF microspheres fabricated using the microfluidic device with flow rates of 0.2 and 0.01 mL/min for continuous phase and disperse phase, respectively was determined to be 73.5% whereas that with flow rates of 0.1 and 0.01 mL/min was measured to be 95.4%. The increase of 29.8% in EE is attributed to the increased average size of the microspheres from 3.0 µm to 52.0 µm. A lower ratio of flow rate of continuous phase to that of disperse phase led to the formation of larger microspheres, quicker solidification of microspheres and thus the observed higher encapsulation efficiency.26 This is in good agreement with a previous report on the effect of microsphere size and encapsulation efficiency.36 Therefore, the produced PPF microspheres having the encapsulation efficiency of 95.4% and well controlled particle size that can enhance the injectability of the microsphere/scaffold composites were selected for the subsequent studies of interaction between Rg1 and PPF microspheres, release behavior and mechanism and angiogenesis activity as explained in detail below. Table 1. Encapsulation Efficiency (EE) of Microspheres Prepared under Different Conditions flow rates (mL/min) preparation method

microfluidic double emulsion

continuous phase

disperse phase

EE (%)

0.20

0.01

73.5±2.3

0.10

0.01

95.4±0.8

n/a

n/a

78.5±1.6


16

Page 17 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3.2 Interaction between Rg1 and PPF microspheres To investigate the presence of ginsenoside Rg1 in the PPF microspheres, the FTIR spectra of Rg1, pure PPF, and Rg1-loaded PPF microspheres were measured and are compared in Figure 5A. In the FTIR spectrum of Rg1 (Figure 5Aa), a few major peaks appear at 3332, 2928, 2875, 1643, 1447, 1369, and 1011 cm -1, attributable to O-H stretching, C-H asymmetric stretching, CH symmetric stretching, C=C stretching/O-H bending, C-H asymmetric deformation, C-H symmetric deformation, and C-O stretching bands, respectively. The FTIR spectrum of PPF (Figure 5Ab) exhibits several major characteristic peaks at 3080, 2985, 2943, 2885, 1713, 1645, 1455, 1250, 1147, and 975 cm-1, attributed to the =CH stretching, CH3 asymmetric stretching, CH2 asymmetric stretching, CH3 symmetric stretching, C=O stretching, C=C stretching, C-H asymmetric deformation, asymmetric C-O-C stretching, symmetric C-O-C stretching, and CH=CH deformation bands, respectively.37-38 After ginsenoside Rg1 was loaded into the PPF microspheres, all the PPF peaks are evident in the spectrum of Rg1-loaded PPF microspheres (Figure 5Ac), confirming the structural stability of PPF during the formation of microspheres. In comparison with the spectrum of PPF (Figure 5Ab), a weak broad peak at 3332 cm-1 and higher peaks at 2985, 2943, and 2885 cm-1 are observed in the spectrum of Rg1-loaded PPF microspheres (Figure 5Ac). The broad O-H peak is ascribed to the presence of alcoholic OH groups of Rg1, which became much weaker after encapsulation due to lower molar fraction of Rg1 in the PPF matrix. The higher ratio of the C-H peaks between 2885 cm-1 and 2985 cm-1 to the C=O peak at 1713 cm-1 is attributed to the higher molar ratio of C-H groups to C=O group. These changes of the spectra indicate successful encapsulation of Rg1 in the PPF matrix. The physical state of the encapsulated Rg1 in the PPF microspheres was further examined by XRD (Figure 5B). As evident in the XRD pattern of Rg1-loaded PPF microspheres (Figure 5Bd), the crystalline peak at 2θ = 5°, which is evident in Figure 5Bc attributed to semicrystalline nature


17


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 37

of Rg1, disappeared after encapsulation within the PPF microspheres fabricated using the microfluidic technique, revealing the amorphous state of Rg1 in the PPF microspheres.26 Therefore, XRD analysis of the microspheres confirms that ginsenoside Rg1 was dispersed molecularly into the PPF matrix with no crystals being present in the Rg1-loaded matrices.39


18

Page 19 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. A) FTIR spectra of (a) ginsenoside Rg1, (b) PPF, and (c) Rg1-loaded PPF microspheres prepared using microfluidic method (flow rates of continuous and disperse phases were 0.10 and 0.01 mL/min, respectively). B) XRD patterns of (a) background, (b) placebo PPF microspheres prepared using microfluidic method, (c) ginsenoside Rg1, and (d) Rg1-loaded PPF microspheres prepared using microfluidic method (flow rates of continuous and disperse phases were 0.10 and 0.01 mL/min, respectively).

As techniques such as XPS can only analyze materials at their surface, XANES spectroscopy provides high energy synchrotron data to determine important details of structure and bonding of absorbing atoms and their immediate surroundings.40 This high resolution technique allows us to probe the interactions between the drug molecules and polymer matrix by tracking the changes in electronic structures (XANES spectra) of carriers before and after drug loading.41 Figure 6a displays the C K-edge XANES spectra of the PPF microspheres with/without Rg1 prepared using the microfluidic technique as well as pure Rg1 for comparison. The spectrum of the pure PPF microspheres (green curve) is characterized by features at 285.6 (“a”), 290.1 (“d”), and 294.5 (“e”) eV. Feature “a” and “e” are attributed to the C=C π* and σ* transitions, while feature “d” is assigned to the C=O π* transition.42-43 The spectrum of the pure Rg1 (blue curve) is a bit more complicated showing features at 286.9 (“b”), 289.1 (“c”), 290.1 (“d”), and 294.5 (“e”) eV. Feature “b” is proposed to come from the C-O-C bond or C-OH π* transition.44 Feature “d” is attributed to the C-H* resonance, while the shoulder “c” at lower energy is assigned to the C-H* resonance of the carbon not bonded to oxygen.42 Feature “e” corresponds to the C-C σ* and C-O σ* resonances.42 Comparing the XANES spectra of the PPF microspheres with/without Rg1 molecules showed that in the spectrum of the Rg1-loaded PPF microspheres (red curve), i) the


19


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 37

intensity of feature “a” drops off by about 25% compared to that of pure PPF microspheres, ii) feature “a” becomes broadened, and iii) a weak shoulder shows up at the lower energy side of feature “d”. The broadening of feature “a” and presence of shoulder “c” in the spectrum of the Rg1-loaded PPF microspheres are due to the contribution from Rg1 molecules (feature “b” and “c”), which further indicates that Rg1 molecules were successfully loaded into the PPF microspheres. The decrease of the intensity of feature “a” relative to the total edge jump could be due to: i) Rg1 molecules contributed to the total edge jump and ii) the decrease of the number of carbon-carbon double bonds in the PPF polymer chain after Rg1 loading. If the Rg1 molecules were physically loaded or sealed in the PPF microspheres without chemical interaction between each other, the Rg1 and PPF both contributed to the total edge jump at the C K-edge, and the weight ratio between Rg1 and PPF would be about 1:3 that resulted in the dropping of the “a” intensity by 25%. Figure 6b shows the linear combination fitting result of the Rg1-loaded PPF microspheres (the weight ratio of Rg1 and PPF is 1:3). The fitting spectrum (black curve) matches well with the experimental one (red curve) except for the intensities of feature “a” and “b”. As shown in the inset of Figure 6b, the intensity of feature “a” in the fitting curve is a bit stronger than that in the experimental one, while feature “b” is weaker, which may be due to chemical interaction between the Rg1 molecules and PPF microspheres. During the Rg1 loading process, some of the carbon-carbon double bonds of PPF may react with the various -OH functional groups from Rg1, forming C-O-C and C-C bonds. After the Rg1 loading process, the decrease of C=C in PPF and the presence of C-O-C bond between Rg1 and PPF lead to the dropping of feature “a” and the rising of feature “b”, respectively.


20

Page 21 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 6. a) C K-edge XANES spectra of pure PPF microspheres, ginsenoside Rg1, and Rg1loaded PPF microspheres prepared using microfluidic method (flow rates of continuous and disperse phases were 0.10 and 0.01 mL/min, respectively). b) Linear fitting of the Rg1-loaded PPF microspheres prepared using microfluidic method. Inset shows the π* resonance.

3.3 Release of Rg1 from delivery system in vitro For comparative purposes, we assayed the in vitro release kinetic profiles of ginsenoside Rg1 from both polydisperse microspheres prepared using the conventional double emulsion technique


21


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 37

and the more monodisperse microspheres fabricated using the microfluidic approach, as shown in Figure 7. In general, Rg1 release behavior for all polydisperse and more monodisperse PPF microspheres exhibited a biphasic profile: i.e. a burst release or initial burst release in the first 2 days and a slow and sustained release over the following 40-day period. The release profiles show that the total cumulative Rg1 release from the more monodisperse microspheres and the polydisperse microspheres are 48.7% and 88%, respectively. In addition, the initial burst release was measured to be 18.5% and 56.5% for the mono and polydisperse microspheres, respectively. As reported by Xu et al.16, smaller particles, due to their larger surface area-to-volume ratio, release drug quicker than larger particles. Moreover, for particles of the same size, the monodisperse microparticles released encapsulated drug more gradually than the polydisperse particles with a lower initial burst rate. Both the slow release and small initial burst release are encouraging for long-term release as required for bone regeneration application. It has also been reported that the viscosity of the polymer solution used in the microsphere fabrication process has a major effect on the initial burst release of entrapped bioactive agent.13 To minimize the effect of viscosity, the PPF microspheres prepared by the two methods were fabricated using the same batch and viscosity of synthesized polymer. Both the smaller initial burst effect and slower release kinetics of the microspheres prepared by the microfluidic method are attributed to homogeneous mixing in the microfluidic device, contributing to a more uniform and even distribution of drug within the interior of the microspheres, which is corroborated by the FTIR, XRD and XANES results. The microspheres prepared by the double emulsion process have drug-rich domains close to the surface of microspheres, attributed to the inhomogeneous forces involved.16 The double emulsion method produced a broad size distribution of particles, particularly a large number of small particles. These will contribute to the larger initial burst and


22

Page 23 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


faster overall kinetics noticed in the polydisperse microparticles, owing to the larger surfacearea-to-volume ratio of the smaller particles.16 In addition, a large amount of Rg1 absorbed on the surface of the microspheres diffused easily into PBS solution resulting in the large initial burst and fast release kinetics.36 Using the microfluidic approach and by incorporating Rg1-loaded PPF microspheres, PPF scaffolds could be rendered angiogenic. It is expected that incorporation of microsphere into microsphere/scaffold composites may result in a negligible burst release and a sustained in vitro release for extended time period.

Figure 7. Release profiles of ginsenoside Rg1 from the more monodisperse PPF microspheres (size: 52.0 µm) prepared with microfluidic device (flow rates of continuous and disperse phases were 0.10 and 0.01 mL/min, respectively) and the polydisperse PPF microspheres prepared using the conventional double emulsion technique.

To understand the release mechanism from the microspheres, we fit the Rg1 release curves in Matlab to determine if the release behavior is controlled by Fickian diffusion, erosion, or a combination of both. Four well-known empirical models were examined, i.e. the Hixon−Crowell,


23


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 37

Korsmeyer−Peppas, Weibull, and first-order models.30-32 According to the r values shown in Table 2, the Weibull model provides the best fit for the Rg1 release experimental data for all the examined delivery systems. The curves of the best fitting and the linear regression lines by the Weibull model are shown in Figure 8. Table 2. Release Model Parameters of Different Carriers for Rg1 at pH 7.4 model 1

First-order Hixon-Crowell

r2

Equation 0.9165

∞

1

0.03839

SSEc

0.793 1.223

0.06173

0.002522

0.303 0.0467

0.06673

0.171

0.944 0.0756

∞

P-MS a

KorsmeyerPeppas

∞

1 Weibull

1

First-order

1

Hixon-Crowell

1

0.3356

0.317

0.985 0.065

∞

0.2468

∞

0.01524

0.4681

0.01409

0.702 0.314 0.512 0.603

∞

M-MSb

KorsmeyerPeppas

2.028 1

Weibull

0.4448

0.903 0.922

∞

1

1.956

0.517

0.926 0.923

∞

a

P-MS: polydisperse microsphere; bM-MS: more monodisperse microsphere;

c

SSE: sum of

squared errors of prediction.


24

Page 25 of 37

0.9 0.8 Weibull model Poly-MS

0.7 0.6 0.5

b)

a)

0.4

.

.

LnLn(1-(1-(Mt/M0)))

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


0.3 0

10

20

30

40

time (days)

.

c)

.

d)

Figure 8. Release profiles (left) and fitting regression lines (right) by Weibull kinetic model for (a, b) Rg1 release from PPF microspheres produced by conventional method and (c, d) Rg1 release from PPF microspheres produced by microfluidic method.

Almost comparable fitting with the Weibull model was observed in the case of using the Korsmeyer−Peppas model for Rg1 release from both systems. A poor fitting was obtained by using Hixson−Crowell and the first-order equations, indicating that they are not proper models for the systems of this study. From Table 2 and Figure 8, the slope of the regression line of the Weibull model is the constant b. According to the literature,45 for values of b lower than 0.75, the


25


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 37

release follows Fickian diffusion, whereas for b > 0.75, other mechanisms such as erosion describe the release better. In this study the results indicate that this value for the polydisperse microspheres and more monodisperse microspheres is 0.317 and 0.517, respectively, as seen in Table 2. Therefore, the diffusion mechanism controlled the release of Rg1 within the studied time which can be described by both the Weibull and Peppas models32, 36 and can be used for longterm drug delivery applications. These results are also consistent with the FTIR, XRD and XANES results, which showed good dispersion of the Rg1 in the PPF polymer matrix.

3.4 Angiogenesis activity of Rg1 Figure 9 displays the tube formation of EA.hy926 cell line seeded in a growth factor-reduced Matrigel-coated 96-well plate. This bioassay compares tube formation in both the presence and absence of ginsenoside Rg1 of different concentrations loaded or released from the PPF microspheres. After seeding for 12-18 h in the absence of angiogenic factor, HUVE cell line formed incomplete and narrow tube-like structures (control sample, Figure 9a); however, in the presence of Rg1 within the dose range of 1-32 μg/mL, elongated and robust capillary-like networks formed, which were organized by a greater number of cells compared with the control indicating the proangiogenesis activity of ginsenoside Rg1, as shown in Figure 9b-g. As seen in Figure 9o, the angiogenesis effect increased generally with increased concentration of Rg1. To study the angiogenesis behavior of ginsenoside Rg1 released from the PPF microspheres, the extract liquid in 2 mL of medium from the 19th day to the 20th day of release was collected. The concentration of Rg1 was measured using UV-Vis spectrophotometry, and diluted to different concentrations of 1-32 µg/mL. It was found that tube formation stimulation activity of ginsenoside Rg1 released from the delivery carriers did not change significantly. This indicates that the examined PPF microsphere delivery systems did not significantly affect the angiogenic


26

Page 27 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


activity of Rg1. Similarly, ginsenoside Rg1 released from the PPF microspheres exhibits similar potency towards the formation of networks of honeycomb-like or tube-like structures in the EA.hy926 cell line, as illustrated in Figure 9i-n. In addition, the angiogenesis effects of Rg1 and the released Rg1 from the microspheres were examined by measuring tube length, number of tubes, number of nodes, number of loops and loop area of these samples, as summarized in Table 3. In general, the angiogenesis effect increased with increasing concentration of Rg1. Hence, Rg1 incorporation into PPF microspheres may result in a slow in vitro release from microsphere/scaffold composites maintaining local in vivo concentrations at angiogenic levels for more adequate time. For that reason, additional consideration to optimize Rg1 pharmacokinetics is mandatory in order to obtain a release profile that could potentially coincide better with the standard rates of formation of tube and bone.46 In addition, it is a complex process to form a network of tubular structures by HUVECs across the surface of a Matrigel substratum combining several elements of attachment, migration, organization, and differentiation. Such complex organizational behavior of HUVECs on Matrigel represents the coordinated activities that are relevant to angiogenesis by endothelial cells. Although true angiogenesis is different from the in vitro Matrigel model, Rg1 is believed to be important to many activities required for vessel formation.47


27


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure

9.

Tube

formation

of

HUVECs

cell

line

Page 28 of 37

cultured

in

different

extract-contained media from PPF microspheres. Effect of ginsenoside Rg1 (before incorporation into PPF microspheres) on tube formation of EA.hy926 cell line: (a) control, (b) 1 µg/mL, (c) 2 µg/mL, (d) 4 µg/mL, (e) 8 µg/mL, (f) 16 µg/mL, (g) and 32 µg/mL ginsenoside Rg1, Rg1induced angiogenic tube formation of EA.hy926 cell line cultured in media contained extract released from Rg1 loaded-PPF microspheres for the the 19th day to 20th day of release at (i) 1 µg/mL, (j) 2 µg/mL, (k) 4 µg/mL, (l) 8 µg/mL, (m) 16 µg/mL, and (n) 32 µg/mL ginsenoside Rg1 compared to (h) blank, and (o) effect of Rg1 concentration on number of branching points for the 19th day to 20th day of release.


28

Page 29 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 3. Quantitative Analysis of Tube Formation Assay by Wimasis Sample

Tube Length [px]a

No. of Tubes b

No. of Nodes c

No. of Loopsd

Loop Area [px]e

a

2635

83

25

1

251

b

3970

97

42

6

6128

c

2557

110

56

16

2402

d

5849

116

65

21

5055

e

2665

105

58

18

1956

f

2455

103

52

10

2416

g

6961

148

83

28

2728

h

3776

89

37

4

6281

i

4307

79

38

8

12574

j

5439

119

65

19

4675

k

5649

105

56

18

7012

l

2533

123

61

13

1378

m

2678

111

58

20

2008

n

2540

112

53

10

2972

a

Tube Length [px]: the length in pixels of the whole tubular structure; bNo. of Tubes: the

number of tubes on the image (a tube is considered to be the part of the tubular structure between two nodes or a node and a loose end.); cNo. of Nodes: the number of nodes (nodes are parts of the skeleton where three or more tubes converge.); dNo. of Loops: the number of loops (a loop is an area of the background enclosed (or almost) by the tubular structure.); eLoop Area [px]: for each loop, the area (number of pixels) enclosed by it is considered as its area.


29


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

4

Page 30 of 37

CONCLUSIONS In this study, ginsenoside Rg1 was encapsulated in PPF microspheres fabricated using both

microfluidic and conventional double emulsion methods, with the microfluidic method generating microspheres of more uniform size. SEM, FTIR, and XRD analyses were employed to examine the morphology of microspheres, chemical stability of the polymer during the formation of microspheres, and physical state of Rg1 in the microspheres. PPF microspheres loaded with/without ginsenoside Rg1 were also characterized by the C K-edge XANES study, confirming that Rg1 was successfully loaded in the PPF microspheres, and the weight ratio between Rg1 and PPF is about 1:3. It is proposed that chemical interaction between the Rg1 and PPF took place via C-O-C and C-C bonds during the Rg1 loading process. The more monodisperse PPF microspheres (50-65 μm) prepared using the microfluidic method exhibited a slower release rate and a smaller initial burst effect compared to the polydisperse microspheres (2-45 μm) produced using the double emulsion method. Moreover, the in vitro release kinetics data followed a Fickian trend with the best fit observed with the Weibull model. In addition, the ginsenoside Rg1 released from the delivery systems showed similar tube formation stimulation effect on human umbilical vein endothelial cell lines to Rg1 before loading into the PPF microspheres, confirming that the microsphere preparation method did not affect the bioactivity of ginsenoside Rg1 in vitro. More monodisperse Rg1-loaded PPF microspheres could be incorporated to the scaffold formulations as an extracellular matrix which can release Rg1 to improve vascularization and thus enhance bone regeneration.


30

Page 31 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ACKNOWLEDGEMENTS The authors thank Dr. Tingjie Li and Prof. Jun Yang, Department of Mechanical and Material Engineering, University of Western Ontario, for assistance with PDMS microfluidic chip preparation. They gratefully thank Ms. Hua Pei, Department of Physiology and Pharmacology, University of Western Ontario, for assistance with the reported bioassay work and Dr. Jeff Dixon, Department of Physiology and Pharmacology, University of Western Ontario, for his donation of umbilical vein cell lines (EA.hy926). They also thank Dr. Mark Biesinger, Surface Science Western, for SEM analysis, Ms. Aneta Borecki, Department of Chemistry, University of Western Ontario, for the powder XRD, and Ms. Dong Zhao, Department of Chemistry, University of Western Ontario, for the XANES measurements. CLS was supported by CFI, NSERC, CHIR, NRC, and the University of Saskatchewan. Financial funding was provided from OGIRC, the Canadian Foundation for Innovation (CFI), NSERC, OIT, OMRI and CRC. ASSOCIATED CONTENT Supporting Information

The Supporting Information is available free of charge on the ACS Publications website at DOI:xxxxxxx/acsbiomaterials.xxxxxxxx. UV-vis spectra and calibration curve of ginsenoside Rg1 solutions (Figure S1) (PDF) AUTHOR INFORMATION Corresponding Author *E-mail: [email protected]; Phone: (519) 661-3466; Fax: (519) 661-3498


31


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 37

REFERENCES (1) Lee, K.-W.; Wang, S.; Fox, B. C.; Ritman, E. L.; Yaszemski, M. J.; Lu, L., Poly (propylene fumarate) bone tissue engineering scaffold fabrication using stereolithography: effects of resin formulations and laser parameters. Biomacromolecules 2007, 8, 1077-1084. (2) Shah, N. J.; Hyder, M. N.; Moskowitz, J. S.; Quadir, M. A.; Morton, S. W.; Seeherman, H. J.; Padera, R. F.; Spector, M.; Hammond, P. T., Surface-Mediated Bone Tissue Morphogenesis from Tunable Nanolayered Implant Coatings. Science Translational Medicine 2013, 5, 191ra83. (3) Hedberg, E. L.; Kroese-Deutman, H. C.; Shih, C. K.; Crowther, R. S.; Carney, D. H.; Mikos, A. G.; Jansen, J. A., In vivo degradation of porous poly (propylene fumarate)/poly (DLlactic- co-glycolic acid) composite scaffolds. Biomaterials 2005, 26, 4616-4623. (4) Goto, K.; Tamura, J.; Shinzato, S.; Fujibayashi, S.; Hashimoto, M.; Kawashita, M.; Kokubo, T.; Nakamura, T., Bioactive bone cements containing nano-sized titania particles for use as bone substitutes. Biomaterials 2005, 26, 6496-6505. (5) Kim, C.; Mahar, A.; Perry, A.; Massie, J.; Lu, L.; Currier, B.; Yaszemski, M. J., Biomechanical evaluation of an injectable radiopaque polypropylene fumarate cement for kyphoplasty in a cadaveric osteoporotic vertebral compression fracture model. Journal of spinal disorders & techniques 2007, 20, 604-609. (6) Kasper, F. K.; Tanahashi, K.; Fisher, J. P.; Mikos, A. G., Synthesis of poly(propylene fumarate). Nature protocols 2009, 4, 518-525. (7) Salarian, M.; Xu, W. Z.; Biesinger, M. C.; Charpentier, P. A., Synthesis and characterization of novel TiO2-poly(propylene fumarate) nanocomposites for bone cementation. Journal of Materials Chemistry B 2014, 2, 5145-5156. (8) Lee, K.-W.; Wang, S.; Yaszemski, M. J.; Lu, L., Physical properties and cellular responses to crosslinkable poly (propylene fumarate)/hydroxyapatite nanocomposites. Biomaterials 2008, 29, 2839-2848. (9) Lalwani, G.; Henslee, A. M.; Farshid, B.; Lin, L.; Kasper, F. K.; Qin, Y.-X.; Mikos, A. G.; Sitharaman, B., Two-Dimensional Nanostructure-Reinforced Biodegradable Polymeric Nanocomposites for Bone Tissue Engineering. Biomacromolecules 2013, 14, 900-909. (10) Mitha, M. K.; Jayabalan, M., Studies on biodegradable and crosslinkable poly(castor oil fumarate)/poly(propylene fumarate) composite adhesive as a potential injectable biomaterial. Journal of materials science. Materials in medicine 2009, 20, 203-211.


32

Page 33 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(11) Hernández, L.; Fernández, M.; Collía, F.; Gurruchaga, M.; Goñi, I., Preparation of acrylic bone cements for vertebroplasty with bismuth salicylate as radiopaque agent. Biomaterials 2006, 27, 100-107. (12) Kempen, D. H.; Lu, L.; Kim, C.; Zhu, X.; Dhert, W. J.; Currier, B. L.; Yaszemski, M. J., Controlled drug release from a novel injectable biodegradable microsphere/scaffold composite based on poly (propylene fumarate). Journal of Biomedical Materials Research Part A 2006, 77, 103-111. (13) Kempen, D. H.; Lu, L.; Zhu, X.; Kim, C.; Jabbari, E.; Dhert, W. J.; Currier, B. L.; Yaszemski, M. J., Development of biodegradable poly (propylene fumarate)/poly (lactic‐co‐glycolic acid) blend microspheres. II. Controlled drug release and microsphere degradation. Journal of Biomedical Materials Research Part A 2004, 70, 293-302. (14) Okushima, S.; Nisisako, T.; Torii, T.; Higuchi, T., Controlled production of monodisperse double emulsions by two-step droplet breakup in microfluidic devices. Langmuir 2004, 20, 9905-9908. (15) Zhang, H.; Tumarkin, E.; Peerani, R.; Nie, Z.; Sullan, R. M. A.; Walker, G. C.; Kumacheva, E., Microfluidic production of biopolymer microcapsules with controlled morphology. Journal of the American Chemical Society 2006, 128, 12205-12210. (16) Xu, Q.; Hashimoto, M.; Dang, T. T.; Hoare, T.; Kohane, D. S.; Whitesides, G. M.; Langer, R.; Anderson, D. G., Preparation of monodisperse biodegradable polymer microparticles using a microfluidic flow‐focusing device for controlled drug delivery. Small 2009, 5, 15751581. (17) Kleinheinz, J.; Stratmann, U.; Joos, U.; Wiesmann, H.-P., VEGF-activated angiogenesis during bone regeneration. Journal of oral and maxillofacial surgery 2005, 63, 1310-1316. (18) Kaigler, D.; Wang, Z.; Horger, K.; Mooney, D. J.; Krebsbach, P. H., VEGF scaffolds enhance angiogenesis and bone regeneration in irradiated osseous defects. Journal of Bone and Mineral Research 2006, 21, 735-744. (19) Villars, F.; Bordenave, L.; Bareille, R.; Amedee, J., Effect of human endothelial cells on human bone marrow stromal cell phenotype: role of VEGF? Journal of cellular biochemistry 2000, 79, 672-685. (20) Leung, K. W.; Cheng, Y.-K.; Mak, N. K.; Chan, K. K.; David Fan, T.; Wong, R. N., Signaling pathway of ginsenoside-Rg1 leading to nitric oxide production in endothelial cells. FEBS letters 2006, 580, 3211-3216. (21) Leung, K. W.; Pon, Y. L.; Wong, R. N.; Wong, A. S., Ginsenoside-Rg1 induces vascular endothelial growth factor expression through the glucocorticoid receptor-related phosphatidylinositol 3-kinase/Akt and β-catenin/T-cell factor-dependent pathway in human endothelial cells. Journal of Biological Chemistry 2006, 281, 36280-36288.


33


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 37

(22) Yue, P. Y.; Mak, N. K.; Cheng, Y. K.; Leung, K. W.; Ng, T. B.; Fan, D. T.; Yeung, H. W.; Wong, R. N., Pharmacogenomics and the Yin/Yang actions of ginseng: anti-tumor, angiomodulating and steroid-like activities of ginsenosides. Chinese medicine 2007, 2, 6. (23) Sengupta, S.; Toh, S.-A.; Sellers, L. A.; Skepper, J. N.; Koolwijk, P.; Leung, H. W.; Yeung, H.-W.; Wong, R. N. S.; Sasisekharan, R.; Fan, T.-P. D., Modulating Angiogenesis. The Yin and the Yang in Ginseng 2004, 110, 1219-1225. (24) Hwisa, N. T.; Katakam, P.; Chandu, B. R.; Adiki, S. K., Solvent Evaporation Techniques as Promising Advancement in Microencapsulation Vedic Research International Biological Medicinal Chemistry 2013, 1, 8-22. (25) Chiellini, F.; Piras, A. M.; Errico, C.; Chiellini, E., Micro/nanostructured polymeric systems for biomedical and pharmaceutical applications. Nanomedicine 2008, 3, 367-393. (26) Samimi, R.; Salarian, M.; Xu, W. Z.; Lui, E. M.; Charpentier, P. A., Encapsulation of Acetyl Ginsenoside Rb1 within Monodisperse Poly (dl-lactide-co-glycolide) Microspheres Using a Microfluidic Device. Industrial & Engineering Chemistry Research 2014, 53, 11333-11344. (27) Lee, J.; Tan, C. Y.; Lee, S.-K.; Kim, Y.-H.; Lee, K. Y., Controlled delivery of heat shock protein using an injectable microsphere/hydrogel combination system for the treatment of myocardial infarction. Journal of Controlled Release 2009, 137, 196-202. (28) Yu, L.-C.; Chen, S.-C.; Chang, W.-C.; Huang, Y.-C.; Lin, K. M.; Lai, P.-H.; Sung, H.W., Stability of angiogenic agents, ginsenoside Rg 1 and Re, isolated from Panax ginseng: in vitro and in vivo studies. International journal of pharmaceutics 2007, 328, 168-176. (29) Chang, C.-H.; Liao, T.-C.; Hsu, Y.-M.; Fang, H.-W.; Chen, C.-C.; Lin, F.-H., A poly (propylene fumarate)–Calcium phosphate based angiogenic injectable bone cement for femoral head osteonecrosis. Biomaterials 2010, 31, 4048-4055. (30) Higuchi, T., Mechanism of sustained‐action medication. Theoretical analysis of rate of release of solid drugs dispersed in solid matrices. Journal of pharmaceutical sciences 1963, 52, 1145-1149. (31) Dash, S.; Murthy, P. N.; Nath, L.; Chowdhury, P., Kinetic modeling on drug release from controlled drug delivery systems. Acta Pol Pharm 2010, 67, 217-23. (32) Wei, Z.; Wang, C.; Liu, H.; Zou, S.; Tong, Z., Facile fabrication of biocompatible PLGA drug-carrying microspheres by O/W pickering emulsions. Colloids and Surfaces B: Biointerfaces 2012, 91, 97-105. (33) Bashir, S.; Rees, J. M.; Zimmerman, W. B., Simulations of microfluidic droplet formation using the two-phase level set method. Chemical Engineering Science 2011, 66, 47334741.


34

Page 35 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(34) Nihant, N.; Schugens, C.; Grandfils, C.; Jérôme, R.; Teyssié, P., Polylactide Microparticles Prepared by Double Emulsion/Evaporation Technique. I. Effect of Primary Emulsion Stability. Pharmaceutical Research 1994, 11, 1479-1484. (35) Khaled, S. M.; Sui, R.; Charpentier, P. A.; Rizkalla, A. S., Synthesis of TiO2-PMMA Nanocomposite: Using Methacrylic Acid as a Coupling Agent. Langmuir 2007, 23, 3988-3995. (36) He, T.; Liang, Q.; Zhang, K.; Mu, X.; Luo, T.; Wang, Y.; Luo, G., A modified microfluidic chip for fabrication of paclitaxel-loaded poly (l-lactic acid) microspheres. Microfluidics and Nanofluidics 2011, 10, 1289-1298. (37) Shung, A. K.; Timmer, M. D.; Jo, S.; Engel, P. S.; Mikos, A. G., Kinetics of poly (propylene fumarate) synthesis by step polymerization of diethyl fumarate and propylene glycol using zinc chloride as a catalyst. Journal of Biomaterials Science, Polymer Edition 2002, 13, 95108. (38) Peter, S. J.; Suggs, L. J.; Yaszemski, M. J.; Engel, P. S.; Mikos, A. G., Synthesis of poly (propylene fumarate) by acylation of propylene glycol in the presence of a proton scavenger. Journal of Biomaterials Science, Polymer Edition 1999, 10, 363-373. (39) Kajjari, P. B.; Manjeshwar, L. S.; Aminabhavi, T. M., Semi-interpenetrating polymer network hydrogel blend microspheres of gelatin and hydroxyethyl cellulose for controlled release of theophylline. Industrial & Engineering Chemistry Research 2011, 50, 7833-7840. (40) Sham, T., Nanoparticles and nanowires: synchrotron spectroscopy studies. International Journal of Nanotechnology 2008, 5, 1194-1246. (41) Guo, X.; Wu, J.; Yiu, Y.-M.; Hu, Y.; Zhu, Y.-J.; Sham, T.-K., Drug–nanocarrier interaction—tracking the local structure of calcium silicate upon ibuprofen loading with X-ray absorption near edge structure (XANES). Physical Chemistry Chemical Physics 2013, 15, 15033-15040. (42) Kikuma, J.; Tonner, B., XANES spectra of a variety of widely used organic polymers at the C K-edge. Journal of electron spectroscopy and related phenomena 1996, 82, 53-60. (43) Dhez, O.; Ade, H.; Urquhart, S., Calibrated NEXAFS spectra of some common polymers. Journal of electron spectroscopy and related phenomena 2003, 128, 85-96. (44) Chuang, C.-H.; Wang, Y.-F.; Shao, Y.-C.; Yeh, Y.-C.; Wang, D.-Y.; Chen, C.-W.; Chiou, J.; Ray, S. C.; Pong, W.; Zhang, L., The Effect of Thermal Reduction on the Photoluminescence and Electronic Structures of Graphene Oxides. Scientific reports 2014, 4, 4525/1-4525/7. (45) Papadopoulou, V.; Kosmidis, K.; Vlachou, M.; Macheras, P., On the use of the Weibull function for the discernment of drug release mechanisms. International journal of pharmaceutics 2006, 309, 44-50.


35


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 37

(46) Kempen, D. H.; Kruyt, M. C.; Lu, L.; Wilson, C. E.; Florschutz, A. V.; Creemers, L. B.; Yaszemski, M. J.; Dhert, W. J., Effect of autologous bone marrow stromal cell seeding and bone morphogenetic protein-2 delivery on ectopic bone formation in a microsphere/poly (propylene fumarate) composite. Tissue Engineering Part A 2008, 15, 587-594. (47) Liang, H.-C.; Chen, C.-T.; Chang, Y.; Huang, Y.-C.; Chen, S.-C.; Sung, H.-W., Loading of a novel angiogenic agent, ginsenoside Rg1 in an acellular biological tissue for tissue regeneration. Tissue engineering 2005, 11, 835-846.


36

Page 37 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table of Contents graphic

Microfluidic Synthesis and Angiogenic Activity of Ginsenoside Rg1-loaded PPF Microspheres

Mehrnaz Salarian1,2, Raziye Samimi2,3, William Z. Xu3, Zhiqiang Wang4, Tsun-Kong Sham4,5, Edmund M. K. Lui2,6, and Paul A. Charpentier*2, 3


37

Microfluidic Synthesis and Angiogenic Activity of ... - ACS Publications

Recommend Documents