Tunable Volumetric Density and Porous Structure ... - ACS Publications

Subscriber access provided by Iowa State University | Library

Article

Tunable Volumetric Density and Porous Structure of Spherical Poly-#-caprolactone Microcarriers, as Applied in Human Mesenchymal Stem Cell Expansion Jian Li, Alan Tin-Lun Lam, Jessica Pei Wen Toh, Shaul Reuveny, Steve Kah-Weng Oh, and William R. Birch Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b00125 • Publication Date (Web): 21 Feb 2017 Downloaded from http://pubs.acs.org on February 22, 2017

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.

Langmuir 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 41

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

Langmuir

Tunable Volumetric Density and Porous Structure of Spherical Poly-ɛ-caprolactone Microcarriers, as Applied in Human Mesenchymal Stem Cell Expansion Jian Li1, Alan Tin-Lun Lam2, Jessica Pei Wen Toh1, Shaul Reuveny2, Steve Kah-Weng Oh2, and William R. Birch1* 1

Institute of Materials Research and Engineering, A*STAR (Agency for Science, Technology

and Research), 2 Fusionopolis Way, Innovis, #08-03, Singapore 138634. 2

Bioprocessing Technology Institute, A*STAR (Agency for Science, Technology and Research),

20 Biopolis Way, #06-01, Singapore 138668.

* To whom correspondence should be addressed:

William R. BIRCH Institute of Materials Research and Engineering Agency for Science, Technology and Research (A*STAR) 2 Fusionopolis Way, Innovis, #08-03, Singapore 138634 Tel: +65-6319-4811, E-mail: [email protected]

1

ACS Paragon Plus Environment

Langmuir

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

Abstract Polymeric microspheres may serve as microcarrier matrices, for the expansion of anchoragedependent stem cells. They require surface properties that both promote initial cell adhesion and the subsequent spreading of cells, which is a prerequisite for successful expansion. When implemented in a three-dimensional culture environment, under agitation, their suspension under low shear rates depends on the microcarriers having a modest negative buoyancy, with a density of 1.02-1.05 g/cm3. Bioresorbable polycaprolactone (PCL), with a density of 1.14 g/cm3, requires a reduction in volumetric density, for the microspheres to achieve high cell viability and yields. Uniform-sized droplets, from solutions of PCL dissolved in dichloromethane (DCM), were generated by a coaxial microfluidic geometry. Subsequent exposure to ethanol rapidly extracted DCM solvent, solidifying the droplets and yielding monodisperse microspheres with a porous structure, which was demonstrated to have tunable porosity and a core-shell configuration. The variation of process parameters, including the molecular weight of PCL, its concentration in DCM, and the ethanol concentration, served to effectively alter the diffusion flux between ethanol and DCM, resulting in a broad spectrum of volumetric densities of 1.04-1.11 g/cm3. The solidified microspheres are generally covered by a smooth thin skin, which provides a uniform cell culture surface and masks their internal porous structure. When coated with cationic polyelectrolyte and extracellular matrix protein, monodisperse microspheres with a diameter of approximately 150 µm and densities ranging from 1.05-1.11 g/cm3, are capable of supporting the expansion of human mesenchymal stem cells (hMSCs). Validation of hMSC expansion was carried out with positive control of commercial Cytodex 3 microcarriers and negative control of uncoated low-density PCL microcarriers. Static culture conditions generated more than 70% cell attachment and similar yields of 6-fold cell expansion on all coated microcarriers, with poor cell

2


Page 2 of 41

Page 3 of 41

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

Langmuir

attachment and growth on the negative control. Under agitation, coated porous microspheres, with a low density of 1.05 g/cm3, achieved robust cell attachment and resulted in high cell yields of 9-fold cell expansion, comparable to those generated by commercial Cytodex 3 microcarriers.

KEYWORDS: microspheres, polymeric microspheres, porous microspheres, microcarriers, microfluidic technology, polycaprolactone, human mesenchymal stem cells

1. INTRODUCTION Polymeric microspheres, specifically those made from biodegradable and biocompatible materials, are a continuing focus of research. They have wide-ranging potential in biomedicine and biotechnology applications, including: immobilization,1,2 affinity chromatography,3,4 drug delivery,5,6 and microcarriers (MCs) for the culture of anchorage-dependent cells, particularly stem cells and tissue specific cells.7 The present article focuses on a bioresorbable material, which is used as a matrix for the expansion of human mesenchymal stem cells (hMSCs). hMSCs are a population of multipotent cells that are defined by their ability to differentiate into cells of the osteogenic, chondrogenic, tendogenic, adipogenic and myogenic lineages.8-11 They possess high self-renewal activity and high ex-vivo expansion ability, thus making them an ideal choice for tissue engineering cell therapies.12,13 However, therapeutic applications of hMSCs require large quantities of cells, which in turn demand robust and scalable cell expansion methods, to meet the clinical demand. Processes that are well suited for efficiently expanding and processing cells in a scalable manner are optimally served by a three-dimensional environment, in a bioreactor under agitation.14-16 Within this scope, defined microcarriers may take the form of spherical polymeric microspheres, requiring two critical attributes: a modest negative buoyancy

3


Langmuir

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 41

and surface properties that are engineered to promote cell adhesion, spreading, and growth. These matrices, when used in a bioreactor under agitation, offer a three-dimensional system for the scalable processing of anchorage-dependent cells, specifically hMSCs for cell therapeutic applications.17,18 MCs have been prepared from a wide variety of polymeric materials,19,20 including polyethylene,21

polystyrene,17,22

polydimethylsiloxane

rubber,23

acrylamide

polymer,24

cellulose,25 and biodegradable polymers, such as chitosan,26 glycosaminoglycans,27 dextran,28 collagen (gelatin),29,30 and polyester.31 Poly-ɛ-caprolactone (PCL) is a bioresorbable polymer that has been used in human implants,32-34 with an emphasis on bone regeneration, where it advantageously generates minimal acidic by-products from its degradation. Its mechanical properties have been leveraged in load-bearing and macroporous scaffold architectures, with a long degradation time.35 Properties of ideal MCs, described in previous review,16 include a smooth surface to allow cell attachment and spreading, as well as a diameter in the range 100-230 µm, in order to allow growth of about 50-400 cells per bead. Their density should be in the range of 1.02-1.05 g/cm3, to allow suspension under low agitation rates. Critically, their surface properties should promote cell attachment (above 80%) after 2 h, as well as cell spreading within 5 h of seeding. A previous study described the fabrication of solid PCL MCs, using microfluidics, to generate microspheres of uniform diameter.36 These MCs bear some properties described above, with a smooth surface and a diameter in the optimal range of 100-200 µm. They are coated with positive charge (Polyl-lysine (PLL)), to allow efficient cell attachment, and ECM protein to allow efficient cell spreading. However they do not have the optimal volumetric density. The high density of PCL (1.14 g/cm3) proscribes their use in cell culture under agitation. Thus, the present study describes

4


Page 5 of 41

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

Langmuir

the fabrication of lower-density’s spherical PCL MCs, with porosity of 50% or greater, corresponding to a density in the range of 1.04-1.07 g/cm3, when immersed in water, for their scalable implementation in stirred-suspension bioreactors. Several reports describe the fabrication of porous polymer microspheres, by means of diverse technologies, which include salt leaching, gas foaming, and phase separation, or core-shell emulsions (water-in-oil-in-water), etc.37,38 A common method used to fabricate porous microspheres is solvent evaporation, which is used in conjunction with double emulsions (waterin-oil-in-water). Several reports implement porous poly(DL-lactide-co-glycolide) (PLGA), PCL and poly(L-lactic acid) (PLA) microspheres to encapsulate proteins or drugs by leveraging a water-in-oil-in-water (w/o/w) emulsion.37-40 However, these microspheres have a low porosity and high density, which engenders issues for their implementation as MCs. Fabrication methods for porous PLGA and PCL microspheres, using ammonium bicarbonate and camphene, respectively, as porogens have also been reported.41-43 Zhang et al. made open porous PLGA and PCL microspheres, presenting a porous surface and an inter-connective pore structure, by combining solvent evaporation and particle leaching.44 Lim et al. prepared highly porous PCL beads with uniform pore structures using an innovative melt-molding particulate-leaching method, with salt particles that serve as molds.45 These microspheres present large-sized open surface pores, which allow the transplanting cells but complicate cell harvesting from the MCs. Although porous polymer microspheres can also be generated by combining a w/o/w emulsion with solvent extraction or osmotic pressure,46-49 the resulting microspheres present a wide size distribution. Microfluidic double emulsification has also been leveraged with solvent diffusion, to fabricate monodisperse ethyl cellulose hollow microcapsules.50 Porous microspheres have also been fabricated from sodium poly(styrene sulfonate) polyelectrolyte, by using microfluidic

5


Langmuir

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 6 of 41

water-in-oil emulsions and solvent extraction.49 These microspheres, fabricated by microfluidics, have a significant advantage in uniform size, ranging from microns to hundreds of microns.36,51,52 The present article draws inspiration from earlier studies,49,50 describing the fabrication of uniform-sized porous PCL microspheres from w/o/w emulsions, formed by microfluidics and subsequently solidified by solvent extraction. The porous microspheres’ surface morphology and inner structure, as well as their size, porosity and overall density, were probed as a function of process parameters: molecular weight of PCL, its concentration in dichloromethane (DCM) solvent, and the ethanol concentration during solvent extraction. The size distribution of porous microspheres was also regulated by changing the flow rate ratio of the continuous and dispersed phases. Phenomena giving rise to different morphologies of the porous microspheres are discussed in terms of the diffusion flux, as described by Fick's first law and the Stokes-Einstein equation. Previous publications have shown that, for the propagation of human pluripotent stem cells and hMSCs on non-porous PCL microspheres under agitation, a coating of

three layers (i.e.

Fibronectin (FN), PLL and FN) is necessary.53 The PLL coating is needed for the attachment of the negatively charged cells to the positively charged surface, while FN is needed to allow cell spreading and growth. Hence, the present study describes the fabrication of solidified porous PCL microspheres with low volumetric density and validates them in a comparison with similar microspheres of higher density, coated with these multilayers to form a suitable extracellular matrix (ECM) for hMSC expansion. These matrices were assessed in static culture conditions and under agitation, with positive and negative control samples. The required surface properties and volumetric density of PCL microspheres were validated, for optimal adhesion and expansion of hMSCs under agitation.

6


Page 7 of 41

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

Langmuir

2. EXPERIMENTAL SECTION 2.1. Materials PCL with an average nominal molecular weight of 10 KDa, 45 KDa and 80 KDa (referred to as PCL10k, PCL45k and PCL80k, respectively), was sourced from Sigma-Aldrich. Polyvinyl alcohol (PVA), with a nominal Mw of 13 ~ 23 KDa, 87 ~ 89% hydrolyzed, was sourced from Sigma-Aldrich. Dichloromethane (DCM) was purchased from J.T.Baker and sodium hydroxide (NaOH) was purchased from Goodrich Chemical Enterprise. Denatured ethanol (99%, Technical grade) was from International Scientific Pte Ltd in Singapore, and absolute ethanol was purchased from Merck. FN from human plasma and Poly-l-lysine hydrobromide (PLL, Mw 70,000-150,000) were bought from Sigma-Aldirch. Cytodex 3, collagen coated dextran-based MCs (positive control), were purchased from GE HealthCare. Non-porous heavy PCL (1.14 g/cm3) were prepared as previously described.36 Pure water was generated by an ELGA PURELAB® Option Q7 Water System. 1× phosphate buffered saline (PBS) was diluted from 10× PBS with pure water. Pure water and PBS were sterilized at 121 oC for 20 min in an autoclave (SANYO Lab Autoclaves M: MLS 3750). PTFE Tubing (Inner diameter: 0.31 mm) was acquired from Sigma-Aldrich. BD PrecisionGlideTM 30G needles (inner diameter: 0.16 mm; outer diameter: 0.31 mm) were from BD (Becton, Dickinson and Company), and 33G Metal Hub needles (inner diameter: 0.11 mm; outer diameter: 0.21 mm) were acquired from Hamilton. Polypropylene (PP) Syringes of 20 ml volume were sourced from B. Braun Medical Inc. and 50 ml volume syringes were from BD. PP tubes of 1.5 ml, 15 ml and 50 ml were acquired from Greiner Bio-One GmbH. A Gemini 88 Dual Syringe Rate pump was purchased from KD Scientific Inc.

7


Langmuir

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

2.2. Microsphere fabrication A coaxial needle/tubing microfluidic device, mounted as described by Li et al.,36 consisted of PTFE tubing and an 33G needle. These feeds were supplied with continuous and dispersed phases, respectively, using individual 20 ml syringes, mounted on a dual-rate syringe pump. The 33G needle was inserted the tubing, thus injecting the dispersed phase into the continuous phase. A continuous phase consisted of 1.5 wt.-% aqueous PVA, generated by dissolving 3.0 g of PVA in 200 ml of pure water at 80 oC and cooling the solution to room temperature, before use. The dispersed phase was prepared from 20 ml aliquots of PCL (PCL10k, PCL45k, and PCL80k) in DCM, at concentrations ranging from 30 to 240 mg/ml. Flow rates of the continuous and dispersed phases (referred to as Qc and Qd, respectively) were 100 and 5-30 µl/min, respectively. As-formed PCL/DCM droplets were exposed to ethanol at room temperature, which rapidly extracted DCM solvent, thus solidifying the PCL/DCM droplets into porous PCL microspheres. The microspheres were then collected and transferred into a 50 ml plastic tube, where they were rinsed twice in 20 ~ 50 ml of absolute ethanol, further extracting residual DCM from the PCL. Residual PVA was removed by dispersal in 40 ~ 50 ml of pure water, followed by rinsing five times. To increase the surface wettability and generate negative surface charge, the PCL microspheres were incubated in 5 M aqueous sodium hydroxide, for 1 h at room temperature. The sterile environment of a biological safety cabinet (NuAireNu-425-400E) was used to rinse these microspheres five times, with autoclaved water, followed with sanitization in 70% ethanol for 30 min. They were then rinsed a further five times in autoclaved water, before twice rinsing in PBS and then storing them in the same. The sanitized porous PCL45k (pPCL45k) microspheres were coated with cationic polyelectrolyte and ECM protein, as described below, to enable their validation as matrices in cell culture assays.

8


Page 8 of 41

Page 9 of 41

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

Langmuir

2.3. Slicing the microspheres In order to probe their inner structure, porous microspheres were sliced using a cryomicrotome (Leica model CM3050 S). Microspheres were dispersed in clear base mold (OCT PEG matrix, Polyethylene glycol, < 5 wt.-%, from Leica Biosystems Richmond, Inc.), which was solidified at -30 oC. These specimens, mounted onto a disc were placed on the sample holder, perpendicular to the knife blade. 70 µm thick sections were sliced and collected in a 50 ml centrifuge tube, at room temperature, followed by adding pure water and rinsing five times, to dissolve the OCT PEG matrix. Sliced microspheres were imaged to characterize their inner morphology. 2.4. Imaging of microspheres A drop of water containing dispersed microspheres was placed on a microscope glass slide and allowed to dry. The microspheres were then imaged under 50× magnification, using an upright microscope (Olympus Corporation, Model BX51TRF). Nikon’s NIS-Elements D 3.00 image analysis software was used to measure the diameter of over 50 microspheres, yielding a geometric mean of diameters (Dg), and its geometric standard deviation (σg). Dried microspheres and sliced microspheres on a glass slide were coated with a thin layer of gold (20-30 nm), sputtered with a JEOL JFC-1200 Fine Coater. Their surface morphology was imaged with a Scanning Electron microscope (SEM): JEOL LV SEM 6360LA, operated at 10 kV. 2.5. Density measurement of dried porous PCL microspheres NaOH-treated porous PCL microspheres were dried in a vacuum oven at room temperature, after being rinsed five times in pure water and twice in absolute ethanol. The density of dry porous PCL microspheres was measured by using Archimedes' principle. A 1.5 ml water volume, in a 1.5 ml Eppendorf tube, was weighed using a calibrated electronic balance, yielding M1. A mark

9


Langmuir

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 41

was placed on the vertical tube, corresponding to the water height. After draining the water and drying the tube, 40 ~ 80 mg of PCL microspheres were added into the tube and weighed, yielding M2. Water was then added into the tube containing the microspheres, to the level of the aforementioned mark and its weight then measured, yielding M3. The volume (V) of the microspheres was derived from: = ( − )/

(1)

where ρw is the density of water. The density (ρPCL-d) of the dry porous PCL microspheres was determined from the equation:

= /( − )

(2)

The average density of each sample was derived from weighing it three times. The microspheres’ average porosity is inferred from: Porosity = /

(3).

2.6. Calculation of the density of the wet porous PCL microspheres in the water The density of the wet porous PCL MCs in water was calculated as follows:



ρ PCL−w = ρ PCL−d + ρ w 1 − 

ρ PCL−d   ρ PCL 

(4)

where ρPCL is the density of PCL, and ρPCL-w is the density of porous PCL microspheres in water (i.e., the inner voids of porous microspheres were filled with water). 2.7. Surface treatment and coating of microspheres PCL microspheres were coated with ECM proteins, to promote the adhesion and spreading of hMSCs and thus enable their culture.53 Sanitized pPCL45k microspheres were coated with three layers, consisting of FN, PLL, and FN, respectively. Three hundred milligrams of porous PCL microspheres, with a surface area of about 120 cm2, were incubated in 1 ml of 240 µg/ml FN solution in PBS in a 1.5 ml Eppendorf tube, for 15 h at room temperature. The microspheres 10


Page 11 of 41

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

Langmuir

were then rinsed twice with PBS and incubated with 1 ml of 120 µg/ml PLL solution in PBS, for 15 h at room temperature. The microspheres were then rinsed twice in PBS and incubated with 1 ml of 240 µg/ml FN in PBS, for 15 h at room temperature. The resulting microspheres, designated as pPCL45k-FN+PLL+FN MCs, were finally rinsed twice with PBS, and then used as matrices for the expansion of MSCs. 2.8. Cell Attachment and Cell Growth Assay Human Wharton’s Jelly MSCs (WJ-1; purchased from PromoCell) were cultivated in a T175 tissue culture flask (Nunc) in a Minimum Essential Medium Eagle Alpha Modification’s (αMEM) media supplemented with 10% fetal bovine serum. MSCs at passage 4-8 were used. Cells were dissociated with 0.25% Trypsin-EDTA (ThermoFisher Scientific) and 0.4 × 105 cells/mL were seeded with 4-5 cells per one microcarrier onto either 300 mg of pPCL45kFN+PLL+FN (with densities: 1.05, 1.06, 1.10, 1.11 g/cm3, respectively), or 300 mg of nonporous PCL-FN+PLL+FN with a density of 1.14 g/cm3, or 300 mg of uncoated pPCL45k (1.05 g/cm3) (negative control), or 80 mg of Cytodex 3 (positive control), respectively, in a 125 ml Erlenmeyer shake flask (Corning), containing 20 mL of α10 medium. The cell culture was maintained for 7 days under agitation, at 110 rpm, in a 37 oC/5% CO2 incubator. For static conditions, cells were incubated in the same environment for 7 days. Cell number and viability was determined using a NC-3000 NucleoCounter (ChemoMetec). Cell attachment was evaluated 2 h after seeding by quantifying viable unattached cells in aliquots of supernatant (1mL), as previously described by Shekaran et al.53 The percentage of cell attachment was calculated by subtracting the unattached cell density from the initial seeded cell density and dividing this by the number of initial cells seeded. Cell fold-expansion was defined as the cell count at day 7, divided by the initial number of cells seeded.

11


Langmuir

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 41

F-Actin staining of 7 days’ old MC cultures was performed by incubating twice-washed cells in PBS, with 1 µL/mL rhodamine phalloidin (Life Technologies) for 2 hours. After washing with PBS, a fluorescent mounting medium with DAPI (Vectashield) was added to cover the samples and incubated for another 1 h, before imaging with a fluorescence microscope (Axiovert 200M, Carl Zeiss). 2.9. Immunophenotypic analysis Harvested hMSCs, after filtering through a 40-µm sieve (BD Biosciences) to remove MCs, were incubated with mouse primary antibodies (purchased from Bio-legend): CD34, CD70, CD90, and CD105, according to the protocols as previously described by Shekaran et al.53

3. RESULTS AND DISCUSSION Li et al. have described the formation of uniform-sized PCL/DCM droplets and subsequent solidification of PCL microspheres by a needle/tubing microfluidic device, using PVA aqueous solution as the continuous phase and PCL/DCM solution as the dispersed phase.36 Exposing a polymer solution to a poor solvent, results in precipitation of the polymer and its ensuing solidification. Ethanol is miscible with DCM and is a poor solvent for PCL. The present report describes the solidification of uniform-sized PCL/DCM droplets in ethanol, as depicted schematically in Figure 1. Solidifying droplets descend by gravity, as DCM is extracted from them in a process driven by the osmotic pressure at the interface between ethanol and PCL/DCM.54,55 As DCM diffuses into ethanol, the latter penetrates the droplets, replacing DCM. PCL ultimately solidifies into a porous structure, arising from the rapid extraction of DCM. Fick's first law describes molecular diffusion across a concentration gradient:

= −

(5)

12


Page 13 of 41

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

Langmuir

where J is the "diffusion flux", representing the flow of matter across a virtual unit area over a unit time interval (mol m−2 s−1), x is distance (normal to the plane) (m), φ is the concentration of the diffusing species (mol/m3); and /! is thus the concentration gradient of the diffusing molecules. D is the diffusion coefficient (m2/s). Its dependence on temperature, viscosity of the liquid, and the size of the diffusing molecules, is defined by the Stokes-Einstein equation: " $

# = %&'(

(6)

where kB is Boltzmann's constant and T is the temperature (K); η is the dynamic viscosity of the liquid (Pa·s), and r is the radius of the diffusing molecules (m).

3.1. Effect of PCL molecular weight PCL solutions in DCM exhibit increasing in viscosity with PCL molecular weight, for the same concentration.36 Hence, PCL molecular weight is expected to influence the formation and morphology of solidified porous PCL microspheres. Three PCL molecular weights (PCL10k, PCL45k, and PCL80k) were used to fabricate porous microspheres. Figure 2a,d,g depict the size distributions of porous microspheres that were fabricated with PCL10k, PCL45k and PCL80k, respectively. The microsphere size distribution narrows with increasing PCL molecular weight. Figure 2b,c depict several PCL10k porous microspheres that have broken into fragments, presumably due to a mechanical fragility. Moreover, pores on the surface of PCL10k microspheres appear larger than those on PCL45k microspheres (Figure 2e,f). Porosity is almost absent at the surface of PCL80k microspheres (Figure 2h,i). The surface of the microspheres also becomes smoother with increasing PCL molecular weight. The latter may be attributed to the surface tension of polymer solutions increasing with molecular weight,56,57 thus contributing to reducing the

porosity and the microspheres’ surface roughness. The diffusion coefficient 13


Langmuir

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 41

decreases with increasing viscosity of the PCL solutions (Equation 6).36 Thus, increasing PCL molecular weight leads to slower mass transfer between DCM and ethanol (Equation 5). In synergy with the increase in PCL/DCM surface tension, this engenders a decrease in the porosity of solidified microspheres with PCL molecular weight. An ancillary observation, from Figure 2c,f,i, is the hollow core formed at the centre of the porous microspheres. The thickness of the shell surrounding this core increases with PCL molecular weight, from approximately one sixth, to one fourth, and two fifths of the diameter, for PCL10k, PCL45k, and PCL80k, respectively. This gives rise to smaller diameter microspheres, with increasing PCL molecular weight, as shown in Figure 2d,g. Given the fragility of porous PCL10k microspheres and the high density of PCL80k microspheres, PCL45k was selected to further examine microsphere formation.

3.2. Influence of flow rates The size of porous PCL45k (pPCL45k) microspheres was influenced by the ratio of flow rates (Qd/Qc). Figure 3a,d,g and Figure 4a,b,c report a size increase with Qd, while part (g) reveals a broad, highly dispersed size distribution. Figures 4c and Figure 3c,f,i present cross-sectional images of an inner morphology that transitions from a porous core-shell structure to a porosity throughout the microsphere volume, as Qd increases, from 5 to 30 µl/min. A prior study similarly demonstrated that as-formed PCL/DCM droplets increase in size with Qd.36 While larger droplet diameter introduces a modest increase in the diffusion distance (x), between the DCM and ethanol phases, the higher number of PCL/DCM droplets per unit volume of ethanol, resulting from an increased Qd, gives rise to a higher, localised concentration of DCM in ethanol. The increased diffusion length (x) reduces the DCM concentration gradient, ⁄! and thus attenuates JDCM, which solidifies microspheres with a slightly lower porosity, as shown in

14


Page 15 of 41

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

Langmuir

Figures 4c and Figure 3c,f,i. In general, Qd has a minor impact on the porosity of the microspheres, from approximately 64% to 57%. Taking into account the microspheres’ size distribution, the the optimum ratio of Qc/Qd is above 5, with a preferred Qd below 20 µl/min, and a Qc of 100 µl/min.

3.3. Effect of ethanol concentration The osmotic pressure at the interface between ethanol and PCL/DCM droplets, as regulated by the ethanol concentration, influences the morphology of porous PCL microspheres.55 Figure 4 shows the dimensions and SEM images of porous PCL microspheres, solidified from 50 mg/ml PCL45k/DCM solution droplets that were precipitated in ethanol with different concentrations. The size of porous PCL microspheres decreases with ethanol concentration, as shown in Figure 4a,d,g,j. The surface of pPCL45k microspheres, collected in 90% ethanol, exhibit significant porosity (Figure 4b). In contrast, pPCL45k microspheres collected in 30-70% ethanol have smoother surfaces and present smaller pores, as shown in Figure 4b,e,h,k. From Fick's first law, a higher ethanol concentration is expected to engender a higher concentration gradient, thus resulting in a higher diffusion flux of ethanol (fethanol) into the PCL/DCM droplets. Assuming that parameters other than the ethanol concentration remain unaltered, with an ethanol concentration ratio of 90%: 70%: 50%: 30%, the ratio of Jethanol can be calculated to be 3: 2.3: 1.7: 1, giving rise to a corresponding ratio of their porosity, measured at 3.5: 3.1: 2.1: 1 (Figure 4c,f,i,l). The correlation of these ratios reflects an accelerated exchange between ethanol and DCM, within the PCL/DCM droplets. The resulting larger pores serve as channels for the exchange between ethanol and DCM, as shown in Figure 4. During the droplet solidifying process, a porous surface skin layer

15


Langmuir

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 41

forms on the microspheres.58,59 The interfacial tension of the droplets, which increases with decreasing ethanol concentration,60 acts so to minimize their surface area, thus potentially contributing to the reduction of their overall diameter.58 Figure 4c,f,i,l depict the inner structures of porous PCL microspheres, solidified by exposure to 90%, 70%, 50%, and 30% ethanol, respectively. Their hollow core increases in size, from quasi-negligible when formed in ≤ 70% ethanol (Figure 4l), to a core-shell structure when formed in 90% ethanol (Figure 4c). Hence, the porosity of porous PCL microspheres decreases with ethanol concentration. Large-sized, lowdensity, and high-porosity PCL microspheres are thus formed by exposing droplets to high ethanol concentration, preferably 70-90%.

3.4. Effect of PCL concentration

Figure 5 displays the size and morphology of pPCL45k microspheres, formed from different concentrations of PCL45k/DCM solution. As shown in Figure 5a,d,g,j,m,p, the diameter of the porous PCL microspheres increases with PCL concentration, which also narrows their size distribution. Higher PCL concentration yields larger microspheres, with higher density, due to the increased PCL mass within each formed droplet. At lower concentrations, of 30 - 75 mg/ml, solidified microspheres generally present a core-shell structure, which diminishes with increasing PCL concentration (Figure 5c,f,i). Moreover, pores at the surface of the microspheres vanish as the PCL/DCM solution concentration increases (Figure 5b,e,h). Microspheres formed from PCL concentrations above 75 mg/ml have few pores. As the PCL solution viscosity increases with concentration,36 DDCM correspondingly decreases, as expressed by Equation (6). Furthermore, the DCM concentration gradient, between the droplets and ethanol, is reduced with increasing PCL concentration. As reported in Table 1, the resulting concurrent reduction in JDCM 16


Page 17 of 41

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

Langmuir

and DDCM with PCL concentration, gives rise to a transition from low-density (high-porosity) porous core-shell microspheres to high-density quasi-solid (low-porosity) microspheres. As the PCL concentration is increased from 30 to 240 mg/ml, the microspheres’ porosity is correspondingly altered from 73% to 20% (Figure 5), with their corresponding density in water thus ranging from 1.04 to 1.11 g/cm3. Given the wide microsphere size distribution at low PCL concentration, PCL concentrations in a range from 30 to 75 mg/ml are most suitable, with 50 and 75 mg/ml selected as the optimal compromise, yielding low-density microspheres with a sufficiently narrow size distribution, preferred as σg ≤1.1. A thin surface layer (Figure 6), which may be described as a “skin”, is observed on all solidified porous microspheres. This smooth outer layer offers a surface that is suitable for cell attachment, spreading, and growth. Its small holes, with diameter typically below 10 µm (Figure 6a), are smaller than the size of hMSCs and effectively prevent hMSCs from accessing the underlying sphere porosity and its hollow core. Figure 6b,c offer an insight into the structure of solidified microspheres. The thin surface skin (indicated by the white arrow in Figure 6b) covers the inner porous structure, which consists primarily of radial pores (indicated by black arrows), that are putatively generated by solvent extraction from the core to the exterior of the microspheres. While PCL/DCM droplets formed by the microfluidic device are uniform, typically with relative standard deviation below 5%,36 Marangoni-induced turbulence may occur during solvent extraction, as the PCL/DCM droplets in aqueous PVA encounter a high concentration of ethanol. This, in turn, may engender collisions between PCL/DCM droplets, which result in either their rupture or coalescence. This starkly amplifies the size distribution of the resulting solidified microspheres. Coalescence may also give rise to large particle sizes, altering the size distribution.

17


Langmuir

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 41

Higher Qd (or lower Qc) decreases the interval between droplets, thus making collisions more likely and engendering a wider microsphere size distribution, as observed for high Qd (Figure 3a,d,g). PCL solution droplets with low viscosity are more easily broken into smaller droplets, subsequently yielding smaller microspheres, after ethanol extraction. For PCL solution droplets with high viscosity, enhanced droplet coalescence may similarly lead to larger microspheres. As the viscosity of PCL solution decreases with concentration,36 microspheres from low concentration PCL45k/DCM exhibit a wide size distribution, putatively arising from the smaller microspheres, generated by ruptured droplets (Figure 5a).

3.5. Expansion of hMSCs The validation of PCL microspheres as matrices for the growth of stem cells was carried out with pPCL45k microspheres of uniform size, with different volumetric densities, both coated and uncoated. hMSC seeding and culture, in both static and agitation conditions, assessed the performance of these matrices. A coating, consisting of three layers of ECM of FN, PLL and FN, was implemented to promote cell adhesion and subsequent spreading.53 Figure 7a-d show microscopy images of hMSCs growing in static and agitated cultures. Figure 7e,f show that the F-actin stained hMSCs spread around skin surface of the microspheres, following 7 days’ growth under agitated conditions. Both hMSC attachment and expansion were assessed under static culture and agitation, as reported in Figure 8. On the negative control of uncoated pPCL45k, cells exhibit low attachment, less than 30%, and insignificant proliferation, in both static and agitation cultures. In static conditions, there is no significant difference in attachment efficiency and cell density for hMSCs grown on the Cytodex 3 positive control and pPCL45k-FN+PLL+FN MCs with different density. Ensuing cell expansion, at 7 days’ culture, is approximately 6-fold,

18


Page 19 of 41

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

Langmuir

which is similar for all densities of microspheres, and the positive control. This indicates that the surface properties of these coated matrices support cell expansion in a similar manner, irrespective of their inner porous structure and their differences in volumetric density. Under agitation, cells adhere poorly (~ 35%) on high-density’s pPCL45k MCs (1.10, 1.11, and 1.14 g/cm3), and the ensuing expansion is only 3-fold at 7 days’ culture. In contrast, pPCL45kFN+PLL+FN MCs with a lower density (1.05 - 1.06 g/cm3) promote high attachment efficiency (~ 80%), comparable to that generated in static conditions, and the ensuing cell expansion at 7 days’ culture is 9-fold, which is significantly higher than in static conditions and comparable to the cell density achieved by the positive control of commercial dextran-based, collagen-coated Cytodex 3 MCs. Thus, the microspheres’ surface properties, when coated with ECM protein and cationic polyelectrolyte, as well as their ability to remain suspended at moderate agitation, play critical roles in achieving high cell attachment and generating high growth yields. Although lower cell growth in static cultures may partially be ascribed to limited nutrient diffusion within the cell/MC aggregates, the data obtained indicate that coated and low density MCs (below 1.06 g/cm3) are preferred matrices for hMSC expansion under agitation. The limited suspension of pPCL45k-FN+PLL+FN MCs with higher density putatively results in collisions that compromise cell attachment efficiency and mitigate the viability of growing cells. Flow cytometry analysis of harvested hMSCs, from various densities of PCL45k-FN+PLL+FN MCs, are shown in Table 1. Cells cultured in static and agitated culture retained their characteristic immunophenotype, as indicated by low expression (< 5%) of marker CD34, and high expression of CD73, CD90 and CD105 markers. However, yields from cultured on high density PCL45k (1.10 and 1.11 g/cm3) in agitated conditions, were insufficient for flow-

19


Langmuir

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 41

cytometry measurements. Briefly, the unaltered expression of surface markers of hMSCs on various MCs indicates that hMSCs cultured on these matrices, under agitation, remain viable.

4. CONCLUSION Porous PCL microspheres, with tunable porosity and volumetric density, were fabricated by combining microfluidic droplet formation with solvent extraction. Their inner porous morphology was tuned by varying PCL molecular weight, its concentration in DCM solution, and the concentration of ethanol used to solidify the microspheres. Their inner structure is regulated by the diffusion flux between extracting ethanol solvent and DCM solvent contained within the PCL/DCM droplets, as determined by Fick's first law and the Stokes-Einstein equation. Low PCL molecular weight, low PCL concentration, and high ethanol concentration enhance the diffusion flux, yielding solidified microspheres with a hollow core surrounded by a porous shell. In contrast, the low diffusion velocity engendered by high PCL molecular weight, high PCL concentration, and low ethanol concentration generated microspheres with smallersized pores, with correspondingly lower porosity and higher density. The surface of the porous microspheres presents a thin outer layer “skin”, with very few pores of small size. Cells grow on this outer layer, without accessing the porous and hollow core interior of the microspheres. It was also found that a narrower size distribution of porous microspheres can be obtained with lower Qd, higher PCL concentration, and higher PCL molecular weight. From empirical data, optimal size, size distribution and porosity of microspheres are generated from 45 kDa of PCL in a 30-75 mg/ml solution, with droplets formed by a Qc/Qd ratio >5 that are subsequently solidified by exposure to 70-90% ethanol, which extracts DCM. Critically, these porous PCL microspheres, coated with ECM protein and cationic polyelectrolyte, are demonstrated to be suitable matrices

20


Page 21 of 41

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

Langmuir

for the expansion of hMSCs. Cell attachment and growth validation confirm the importance of coating the PCL microspheres and achieving a low volumetric density, well below 1.1 g/cm3, which enables high cell yields in an environment under agitation. A decreased volumetric density (≤1.06 g/cm3) allows their suspension at moderate agitation rates and does not significantly compromise cell viability and growth. This is demonstrated by fold-expansions comparable to benchmarking against non-bioresorbable, commercial MCs.

ACKNOWLEDGMENTS The authors would like to acknowledge funding from the Joint Council Office (Singapore, grant no. 1334i00054), as well as support from the Institute of Materials Research and Engineering and the Bioprocessing Technology Institute of the Agency for Science, Technology and Research (A*STAR), Singapore.

REFERENCES (1)

Basinska, T.; Slomkowski, S. The Direct Determination of Protein-concentration for

Proteins Immobilized on Polystyrene Microspheres. J. Biomater. Sci., Polym. Ed. 1991, 3 (2), 115-125. (2)

Zhang, J.; Wang, D.; Pan, J. C.; Wang, J.; Zhao, H. B.; Li, Q.; Zhou, X. H. Efficient

Resveratrol Production by Immobilized Beta-Glucosidase on Cross-linked Chitosan Microsphere Modified by L-lysine. J. Mol. Catal. B-Enzym. 2014, 104, 29-34. (3)

Baruch-Teblum, E.; Mastai, Y.; Landfester, K. Miniemulsion Polymerization of

Cyclodextrin Nanospheres for Water Purification from Organic Pollutants. Eur. Polym. J. 2010, 46 (8), 1671-1678.

21


Langmuir

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 22 of 41

Margel, S. Affinity Separation with Polyaldehyde Microsphere Beads. J. Chromatogr. A

1989, 462 (0), 177-189. (5)

Chang, L.; Peng, C.; Chiang, P.; Wang, H.; Luo, T. Using the Biodegradable PLGA

Microsphere to Evaluated a New Drug Delivery System for the Brachytherapy of Hepatoma in Animal Model. Eur. J. Nucl. Med. Mol. Imaging 2013, 40, S247-S247. (6)

Anindita, M.; Vikas, P. Drug Delivery: Techniques for Polymeric Microsphere

Preparation. Res. J. Biotechnol. 2007, 2 (4), 58-63. (7)

Silva, G. A.; Coutinho, O. P.; Ducheyne, P.; Reis, R. L. Materials in Particulate Form for

Tissue Engineering. 2. Applications in Bone. J. Tissue Eng. Regen. Med. 2007, 1 (2), 97-109. (8)

Pittenger, M. F.; Mackay, A. M.; Beck, S. C.; Jaiswal, R. K.; Douglas, R.; Mosca, J. D.;

Moorman, M. A.; Simonetti, D. W.; Craig, S.; Marshak, D. R. Multilineage Potential of Adult Human Mesenchymal Stem Cells. Science 1999, 284 (5411), 143-147. (9)

Beyer Nardi, N.; Meirelles, L. D. Mesenchymal Stem Cells: Isolation, in vitro Expansion

and Characterization. Stem Cells 2006, 174, 249-282. (10)

Brighton, C. T.; Hunt, R. M. Early Histological and Ultrastructural-changes in Medullary

Fracture Callus. J. Bone Joint Surg. Am. 1991, 73A (6), 832-847. (11)

Brighton, C. T.; Hunt, R. M. Early Histologic and Ultrastructural Changes in

Microvessels of Periosteal Callus. J. Orthop. Trauma 1997, 11 (4), 244-253. (12)

Lin, X. X.; Huang, J.; Shi, Y.; Liu, W. Tissue Engineering and Regenerative Medicine in

Applied Research: A Year in Review of 2014. Tissue Eng. Part B Rev. 2015, 21 (2), 177-186. (13)

Abdallah, B. M.; Kassem, M. The Use of Mesenchymal (Skeletal) Stem Cells for

Treatment of Degenerative Diseases: Current Status and Future Perspectives. J. Cell. Physiol. 2009, 218 (1), 9-12.

22


Page 23 of 41

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

Langmuir

(14)

Schnitzler, A. C.; Verma, A.; Kehoe, D. E.; Jing, D.; Murrell, J. R.; Der, K. A.; Aysola,

M.; Rapiejko, P. J.; Punreddy, S.; Rook, M. S. Bioprocessing of Human Mesenchymal Stem/stromal Cells for Therapeutic Use: Current Technologies and Challenges. Biochem. Eng. J. 2016, 108, 3-13. (15)

Nienow, A. W.; Hewitt, C. J.; Heathman, T. R. J.; Glyn, V. A. M.; Fonte, G. N.; Hanga,

M. P.; Coopman, K.; Rafiq, Q. A. Agitation Conditions for the Culture and Detachment of hMSCs from Microcarriers in Multiple Bioreactor Platforms. Biochem. Eng. J. 2016, 108, 24-29. (16)

Chen, A. K.-L.; Reuveny, S.; Oh, S. K. W. Application of Human Mesenchymal and

Pluripotent Stem Cell Microcarrier Cultures in Cellular Therapy: Achievements and Future Direction. Biotechnol. Adv. 2013, 31 (7), 1032-1046. (17)

Heng, B. C.; Li, J.; Chen, A. K.-L.; Reuveny, S.; Cool, S. M.; Birch, W. R.; Oh, S. K.-W.

Translating Human Embryonic Stem Cells from 2-Dimensional to 3-Dimensional Cultures in a Defined Medium on Laminin- and Vitronectin-Coated Surfaces. Stem Cells Dev. 2012, 21 (10), 1701-1715. (18)

Lam, A. T.; Chen, A. K.; Li, J.; Birch, W. R.; Reuveny, S.; Oh, S. K. Conjoint

Propagation and Differentiation of Human Embryonic Stem Cells to Cardiomyocytes in a Defined Microcarrier Spinner Culture. Stem Cell Res. Ther. 2014, 5 (5), 110. (19)

Malda, J.; Frondoza, C. G. Microcarriers in the Engineering of Cartilage and Bone.

Trends Biotechnol. 2006, 24 (7), 299-304. (20)

Martin, Y.; Eldardiri, M.; Lawrence-Watt, D. J.; Sharpe, J. R. Microcarriers and Their

Potential in Tissue Regeneration. Tissue Eng. Part B Rev. 2011, 17 (1), 71-80.

23


Langmuir

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

(21)

Page 24 of 41

Malda, J.; Kreijveld, E.; Temenoff, J. S.; van Blitterswijk, C. A.; Riesle, J. Expansion of

Human Nasal Chondrocytes on Macroporous Microcarriers Enhances Redifferentiation. Biomaterials 2003, 24 (28), 5153-5161. (22)

Lam, A. T.-L.; Li, J.; Chen, A. K.-L.; Reuveny, S.; Oh, S. K.-W.; Birch, W. R. Cationic

Surface Charge Combined with Either Vitronectin or Laminin Dictates the Evolution of Human Embryonic Stem Cells/Microcarrier Aggregates and Cell Growth in Agitated Cultures. Stem Cells Dev. 2014, 23 (14), 1688-1703. (23)

Denkbas, E. B.; Hoffman, A. S.; Piskin, E. Silicone-based Microcarriers - Preparation

and BHK Cell-culture. Chem. Eng. J. Biochem. Eng. J. 1995, 58 (1), 65-70. (24)

Cadic, C.; Baquey, C.; Dupuy, B. Inverted Microcarriers for Cell-cultures Made by

Polymerization of Shells Around Agarose Microspheres in a Noncytotoxic Procedure. Polym. J. 1991, 23 (8), 933-937. (25)

Reuveny, S.; Silberstein, L.; Shahar, A.; Freeman, E.; Mizrahi, A. DE-52 And DE-53

Cellulose Microcarriers. 1. Growth of Primary and Estabished Anchorage-dependent Cells. In Vitro Cell Dev. B 1982, 18 (2), 92-98. (26)

Chen, X.-G.; Liu, C.-S.; Liu, C.-G.; Meng, X.-H.; Lee, C. M.; Park, H.-J. Preparation and

Biocompatibility of Chitosan Microcarriers as Biomaterial. Biochem. Eng. J. 2006, 27 (3), 269274. (27)

Foran, D. J.; Cahn, F.; Eikenberry, E. F. Assessment of Cell-proliferation on Porous

Microcarriers by Means of Image-analysis. Anal. Quant. Cytopathol. Histpathol. 1991, 13 (3), 215-222.

24


Page 25 of 41

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

Langmuir

(28)

Bardouille, C.; Lehmann, J.; Heimann, P.; Jockusch, H. Growth and Differentiation of

Permanent and Secondary Mouse Myogenic Cell Lines on Microcarriers. Appl. Microbiol. Biotechnol. 2001, 55 (5), 556-562. (29)

Doctor, J. S.; Salvaterra, S.; Vitrant, D.; Azari, K.; Ihnat, M. A.; Hollinger, J. O.;

Campbell, P. Using Collagen Microcarriers to Deliver Cells for Bone Tissue Engineering. Dev. Biol. 2001, 235 (1), 245-245. (30)

Tielens, S.; Declercq, H.; Gorski, T.; Lippens, E.; Schacht, E.; Cornelissen, M. Gelatin-

based Microcarriers as Embryonic Stem Cell Delivery System in Bone Tissue Engineering: An in-vitro Study. Biomacromolecules 2007, 8 (3), 825-832. (31)

Murueva, A. V.; Shershneva, A. M.; Shishatskaya, E. I.; Volova, T. G. The Use of

Polymeric Microcarriers Loaded with Anti-Inflammatory Substances in the Therapy of Experimental Skin Wounds. Bull. Exp. Biol. Med. 2014, 157 (5), 597-602. (32)

Woodruff, M. A.; Hutmacher, D. W. The Return of a Forgotten Polymer—

Polycaprolactone in the 21st century. Prog. Polym. Sci. 2010, 35 (10), 1217-1256. (33)

Dash, T. K.; Konkimalla, V. B. Poly-є-caprolactone Based Formulations for Drug

Delivery and Tissue Engineering: A Review. J. Control. Release 2012, 158 (1), 15-33. (34)

Baji, A.; Wong, S. C.; Srivatsan, T. S.; Njus, G. O.; Mathur, G. Processing

Methodologies for Polycaprolactone-Hydroxyapatite Composites: A Review. Mater. Manuf. Processes 2006, 21 (2), 211-218. (35)

Lam, C. X. F.; Savalani, M. M.; Teoh, S. H.; Hutmacher, D. W. Dynamics of in vitro

Polymer Degradation of Polycaprolactone-based Scaffolds: Accelerated versus Simulated Physiological Conditions. Biomed. Mater. 2008, 3 (3), 15.

25


Langmuir

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

(36)

Page 26 of 41

Li, J.; Lam, A. T.-L.; Toh, J. P. W.; Reuveny, S.; Oh, S. K.-W.; Birch, W. R. Fabrication

of Uniform-sized Poly-epsilon-caprolactone Microspheres and Their Applications in Human Embryonic Stem Cell Culture. Biomed. Microdevices 2015, 17 (6), 105. (37)

Crotts, G.; Park, T. G. Preparation of Porous and Nonporous Biodegradable polymeric

Hollow Microspheres. J. Control. Release 1995, 35 (2-3), 91-105. (38)

Han, K.; Lee, K. D.; Gao, Z. G.; Park, J. S. Preparation and Evaluation of Poly(L-lactic

acid) Microspheres Containing rhEGF for Chronic Gastric Ulcer Healing. J. Control. Release 2001, 75 (3), 259-269. (39)

Yan, C. H.; Resau, J. H.; Hewetson, J.; West, M.; Rill, W. L.; Kende, M. Characterization

and Morphological Analysis of Protein-loaded Poly(lactide-co-glycolide) Microparticles Prepared by Water-in-oil-in-water Emulsion Technique. J. Control. Release 1994, 32 (3), 231241. (40)

Coccoli, V.; Luciani, A.; Orsi, S.; Guarino, V.; Causa, F.; Netti, P. A. Engineering of

Poly(epsilon-caprolactone) Microcarriers to Modulate Protein Encapsulation Capability and Release Kinetic. J. Mater. Sci.-Mater. Med. 2008, 19 (4), 1703-1711. (41)

Kim, T. K.; Yoon, J. J.; Lee, D. S.; Park, T. G. Gas Foamed Open Porous Biodegradable

Polymeric Microspheres. Biomaterials 2006, 27 (2), 152-159. (42)

Shin, U. S.; Park, J.-H.; Hong, S.-J.; Won, J.-E.; Yu, H.-S.; Kim, H.-W. Porous

Biomedical Composite Microspheres Developed for Cell Delivering Scaffold in Bone Regeneration. Mater. Lett. 2010, 64 (20), 2261-2264. (43)

Jin, G.-Z.; Park, J.-H.; Seo, S.-J.; Kim, H.-W. Dynamic Cell Culture on Porous

Biopolymer Microcarriers in a Spinner flask for Bone Tissue Engineering: a Feasibility Study. Biotechnol. Lett. 2014, 36 (7), 1539-1548.

26


Page 27 of 41

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

Langmuir

(44)

Zhang, Q. C.; Tan, K.; Ye, Z. Y.; Zhang, Y.; Tan, W. S.; Lang, M. D. Preparation of

Open Porous Polycaprolactone Microspheres and Their Applications as Effective Cell Carriers in Hydrogel System. Mater. Sci. Eng. C Mater. Biol. Appl. 2012, 32 (8), 2589-2595. (45)

Lim, S. M.; Lee, H. J.; Oh, S. H.; Kim, J. M.; Lee, J. H. Novel Fabrication of PCL Porous

Beads for Use as an Injectable Cell Carrier System. J. Biomed. Mater. Res. B Appl. Biomater. 2009, 90B (2), 521-530. (46)

Jeyanthi, R.; Thanoo, B. C.; Metha, R. C.; DeLuca, P. P. Effect of Solvent Removal

Technique on the Matrix Characteristics of Polylactide/glycolide Microspheres for Peptide Delivery. J. Control. Release 1996, 38 (2-3), 235-244. (47)

Jang, T.-S.; Lee, E.-J.; Kim, H.-E.; Koh, Y.-H. Hollow Porous Poly(epsilon-

caprolactone) Microspheres by Emulsion Solvent Extraction. Mater. Lett. 2012, 72, 157-159. (48)

Zhang, G.-H.; Hou, R.-X.; Zhan, D.-X.; Cong, Y.; Cheng, Y.-J.; Fu, J. Fabrication of

Hollow Porous PLGA Microspheres for Controlled Protein Release and Promotion of Cell Compatibility. Chinese Chemical Letters 2013, 24 (8), 710-714. (49)

Watanabe, T.; Lopez, C. G.; Douglas, J. F.; Ono, T.; Cabral, J. T. Microfluidic Approach

to the Formation of Internally Porous Polymer Particles by Solvent Extraction. Langmuir 2014, 30 (9), 2470-2479. (50)

Liu, L.; Yang, J.-P.; Ju, X.-J.; Xie, R.; Yang, L.; Liang, B.; Chu, L.-Y. Microfluidic

Preparation of Monodisperse Ethyl Cellulose Hollow Microcapsules with Non-toxic Solvent. J. Colloid Interface Sci. 2009, 336 (1), 100-106. (51)

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

27


Langmuir

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 28 of 41

Microparticles Using a Microfluidic Flow-Focusing Device for Controlled Drug Delivery. Small 2009, 5 (13), 1575-1581. (52)

Serra, C. A.; Chang, Z. Microfluidic-assisted Synthesis of Polymer Particles. Chem. Eng.

Technol. 2008, 31 (8), 1099-1115. (53)

Shekaran, A.; Lam, A.; Sim, E.; Jialing, L.; Jian, L.; Wen, J. T. P.; Chan, J. K. Y.;

Choolani, M.; Reuveny, S.; Birch, W.; Oh, S. Biodegradable ECM-coated PCL Microcarriers Support Scalable Human Early MSC Expansion and in vivo bone formation. Cytotherapy 2016, 18, 1332–1344. (54)

Dittrich, M.; Hampl, J.; Soukup, F. Branched Oligoester Microspheres Fabricated by a

Rapid Emulsion Solvent Extraction Method. J. Microencapsul. 2000, 17 (5), 587-598. (55)

Lee, J.; Oh, Y. J.; Lee, S. K.; Lee, K. Y. Facile Control of Porous Structures of Polymer

Microspheres Using an Osmotic Agent for Pulmonary Delivery. J. Control. Release 2010, 146 (1), 61-67. (56)

Legrand, D. G.; Gaines, G. L. Molecular Weight Dependence of Polymer Surface

Tension. J. Colloid Interface Sci. 1969, 31 (2), 162-167. (57)

Sauer, B. B.; Dee, G. T. Molecular-weight and Temperature-dependence of Polymer

Surface-tension-comparison of Experiment with Theory. Macromolecules 1991, 24 (8), 21242126. (58)

Li, W. I.; Anderson, K. W.; DeLuca, P. P. Kinetic and Thermodynamic Modeling of the

Formation of Polymeric Microspheres Using Solvent Extraction/evaporation Method. J. Control. Release 1995, 37 (3), 187-198. (59)

Yang, Y. Y.; Chia, H. H.; Chung, T. S. Effect of Preparation Temperature on the

Characteristics and Release Profiles of PLGA Microspheres Containing Protein Fabricated by

28


Page 29 of 41

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

Langmuir

Double-emulsion Solvent Extraction/evaporation Method. J. Control. Release 2000, 69 (1), 8196. (60)

Sawalha, H.; Fan, Y. X.; Schroen, K.; Boom, R. Preparation of Hollow Polylactide

Microcapsules through Premix Membrane Emulsification-Effects of Nonsolvent Properties. J. Membr. Sci. 2008, 325 (2), 665-671.

29


Langmuir

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 30 of 41

Table 1. Density of porous PCL microspheres, and immunophenotypic flow cytometry analysis of hMSC growth on the corresponding porous microspheres, from the expression of MSC markers CD34, CD73, CD90, and CD105. Porous PCL microspheres, coated with three layers of FN+PLL+FN, were fabricated from 50, 75, 140, and 170 mg/ml PCL45k/DCM solutions, respectively, and collected in 90% ethanol. Qd and Qc were 5 and 100 µl/min, respectively. Concentration of PCL45k/DCM (mg/ml)

50

75

140

170

Diameter Density of wet pPCL45k microspheres in water (g/cm3)

150 ± 10 1.05 ± 0.004

140 ± 7 1.06 ± 0.004

145 ± 7 1.10 ± 0.01

155 ± 5 1.11 ± 0.01

Prior cell seeding (Day 0)

Flow cytometry (expression %)

Static (Day 7)

CD34

0.38+0.2

CD73

98.4+1.9

CD90

95.9+2.3

CD105

94.2+1.5

CD34 CD73 CD90 CD105 CD34

0.36±0.1 98.6±1.2 83.4±1.7 87.1±2.9 1.54±0.3

1.25±0.2 97.1±1.5 85.4±1.7 92.1±1.8 1.97±0.3

1.58±0.7 99.6±1.6 99.3±1.5 95.3±2.1

1.54±0.6 99.8±1.7 96.7±2.5 92.9±2.3

CD73 89.4±1.6 98.2±1.6 Null* CD90 91.0±1.9 98.0±1.4 CD105 87.0±2.6 90.7±2.4 * No cell marker data for 140 and 170mg/mL PCL45kDCM under agitated cultures since no cell Agitation (Day 7)

growth was achieved under these conditions.

30


Page 31 of 41

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

Langmuir

Figure Captions Figure 1. Schematic of porous PCL microspheres, formed from PCL/DCM droplets by ethanol extraction.

Figure 2. Size distributions, derived from the inserted SEM images of PCL porous microspheres, which were solidified in 90% ethanol. Microspheres were formed from 50 mg/ml solution of PCL10k (a, b, c), PCL45k (d, e, f), PCL 80k (g, h, i). Qc and Qd were 100 and 5 µl/min, respectively.

Figure 3. Size distributions and SEM images of porous PCL45k microspheres, formed from a 50 mg/ml PCL45k/DCM solution, and solidified in 90% ethanol. Qc was maintained at 100 µl/min, while Qd was: (a, b, c) 10 µl/min, (d, e, f) 20 µl/min, and (g, h, i) 30 µl/min. Figure 4. Size distributions and SEM images of solidified PCL45k porous microspheres, formed from a 50 mg/ml PCL45k/DCM solution. Porous microspheres were solidified in ethanol, with concentrations of: (a, b, c) 90%, (d, e, f) 70%, (g, h, i) 50%, and (j, k, l) 30%. The SEM images of their corresponding cross-sections are depicted in (c), (f), (i), and (l), respectively. Qc and Qd were 100 and 5 µl/min, respectively.

Figure 5. Size distributions, derived from SEM images of the porous PCL45k microspheres, solidified in 90% ethanol. The PCL/DCM solution concentrations were (a, b, c) 30 mg/ml, (d, e, f) 50 mg/ml, (g, h, i) 75 mg/ml, (j, k, l) 140 mg/ml, (m, n, o) 170 mg/ml, and (p, q, r) 240 mg/ml. Qc and Qd were 100 and 5 µl/min, respectively. Figure 6. SEM images of solidified porous PCL microspheres, solidified in 90% ethanol. Spheres were formed from (a) 50 mg/ml and (b, c) 240 mg/ml of PCL/DCM solution, with Qc = 100 31


Langmuir

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

µl/min and Qd = 5 µl/min, respectively. Image (b) is the magnification of Figure 5r. The white arrow in part (b) points to a thin skin layer on the surface, while the black arrows in parts (b) and (c) highlight the underlying porous structures.

Figure 7. Photographs of hMSCs attached on pPCL45k-FN+PLL+FN MCs, following 7 days’ expansion. Parts (a, c, e, f): pPCL45k MCs, with a low density of 1.05 g/cm3, were prepared from 50 mg/ml PCL45k/DCM solution. Parts (b, d): pPCL45k MCs, with a high density of 1.11 g/cm3, were prepared from 170 mg/ml PCL45k/DCM solution. Cells were proliferated under static conditions (a, b) and in an environment under agitation (c, d, e, f), respectively. Arrows point to hMSCs growing on the microspheres. (e, f) Representative microscopy images of FActin stained hMSCs from (c), and (f) F-Actin Stained hMSCs spreading on the surface of MCs.

Figure 8. hMSC growth on pPCL45k MCs with low and high density under static culture and agitation. (a) Percentage of cell attachment on the MCs after being seeded for 2 h, and (b) Cell density following 7 days’ expansion. The pPCL45k MCs, coated with three layers of FN+PLL+FN, were fabricated from 50, 75, 140, and 170 mg/ml PCL45k/DCM solutions and solidified in 90% ethanol, and have a density of 1.05, 1.06, 1.01 and 1.11 g/cm3, respectively. Qd and Qc were 5 and 100 µl/min, respectively. Cytodex 3 and non-porous PCL-FN+PLL+FN MCs were used as positive control, while uncoated pPCL45k MCs are served as negative controls. ** indicates p < 0.01.

32


Page 32 of 41

Page 33 of 41

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

Langmuir

Figure 1

33


Langmuir

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 2

34


Page 34 of 41

Page 35 of 41

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

Langmuir

Figure 3

35


Langmuir

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

36


Page 36 of 41

Page 37 of 41

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

Langmuir

Figure 5

37


Langmuir

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

38


Page 38 of 41

Page 39 of 41

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

Langmuir

Figure 7

39


Langmuir

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 8

40


Page 40 of 41

Page 41 of 41

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

Langmuir

Table of Contents

41


Tunable Volumetric Density and Porous Structure ... - ACS Publications

Recommend Documents