micromachines

Article A 3D Printed Self-Sustainable Cell-Encapsulation Device for Periocular Transplant-Based Treatment of Retinal Degenerative Diseases

Hideto Kojima 1, Bibek Raut 1 , Li-Jiun Chen 1, Nobuhiro Nagai 2, Toshiaki Abe 2 and Hirokazu Kaji 1,3,*

1 Department of Finemechanics, Graduate School of Engineering, Tohoku University, 6-6-01 Aramaki, Aoba-ku, Sendai 980-8579, Japan; [email protected] (H.K.); [email protected] (B.R.); [email protected] (L.-J.C.) 2 Division of Clinical , United Centers for Advanced Research and Translational Medicine (ART), Tohoku University Graduate School of Medicine, 2-1 Seiryo, Aoba-ku, Sendai 980-8575, Japan; [email protected] (N.N.); [email protected] (T.A.) 3 Department of Biomedical Engineering, Graduate School of Biomedical Engineering, Tohoku University, 6-6-01 Aramaki, Aoba-ku, Sendai 980-8579, Japan * Correspondence: [email protected]; Tel.: +81-22-795-4249

 Received: 27 March 2020; Accepted: 18 April 2020; Published: 21 April 2020 

Abstract: Self-sustainable release of brain-derived neurotrophic factor (BDNF) to the retina using minimally invasive cell-encapsulation devices is a promising approach to treat retinal degenerative diseases (RDD). Herein, we describe such a self-sustainable drug delivery device with human retinal pigment epithelial (ARPE-19) cells (cultured on coated polystyrene (PS) sheets) enclosed inside a 3D printed semi-porous capsule. The capsule was 3D printed with two photo curable : triethylene glycol dimethacrylate (TEGDM) and dimethylacrylate (PEGDM). The capsule’s semi-porous membrane (PEGDM) could serve three functions: protecting the cells from body’s by limiting diffusion (5.97 0.11%) of large molecules like ± immunoglobin G (IgG)(150 kDa); helping the cells to survive inside the capsule by allowing diffusion (43.20 2.16%) of small molecules (40 kDa) like oxygen and necessary nutrients; and helping in the ± treatment of RDD by allowing diffusion of cell-secreted BDNF to the outside environment. In vitro results showed a continuous BDNF secretion from the device for at least 16 days, demonstrating future potential of the cell-encapsulation device for the treatment of RDD in a minimally invasive and self-sustainable way through a periocular transplant.

Keywords: retinal degenerative disease; cell-encapsulation device; periocular implant; growth factors; brain-derived neurotrophic factor (BDNF); cell sheet engineering; 3D printing; minimally invasive device

1. Introduction Retinal degenerative diseases (RDD), such as age-related macular degeneration (AMD) and retinitis pigmentosa (RP), causes progressive damage to the photoreceptor cells of the retina leading to gradual visual decline [1]. Although no permanent cure or prosthetic exists to date, cell culture and animal experiments done with tropic factors, such as brain-derived neurotrophic factor (BDNF) and ciliary neurotrophic factor (CNTF), have shown that they can revive the damaged photoreceptor cells [2–4]. However, their delivery to the retina is very challenging [5,6]. For instance, intravenous injection cannot deliver the required amount of BDNF to the retina because BDNF has a very short half-life in blood (0.92 min) [7], and it is impermeable to the blood-retinal barrier [8]. Likewise, topical

Micromachines 2020, 11, 436; doi:10.3390/mi11040436 www.mdpi.com/journal/micromachines Micromachines 2020, 11, 436 2 of 12

Micromachines 2020, 11, x 2 of 12 installation is equally ineffective due to low permeability through multi-cellular cornea and sclera [9,10]. Moreover,[10]. Moreover, intravitreal intravitreal injection injection is highly is invasive highly invasive during long during term long treatment term treatment that requires that periodic requires pokingperiodic of poking the eyeball of the which eyeball can which risk infectioncan risk infection [9]. Although [9]. Although minimally minimally invasive invasive delivery delivery of drugs of throughdrugs through the blood-retina the blood barrier-retina usingbarrier focused using focused ultrasound ultrasoun [11] hasd been[11] has proposed, been proposed, a minimally a minimally invasive wayinvasive of sustained way of sustained and localized and druglocalized delivery drug is delivery desirable. is desirable. WeWe havehave previouslypreviously developeddeveloped transscleraltransscleral (periocular)(periocular) implantsimplants asas aa minimallyminimally invasiveinvasive wayway toto deliver deliver drugs drugs to to the the retina retina [12[12–15–15].]. TheseThese implants implants are are generally generally placed placed outside outside the the eyeball eyeball (subconjunctival,(subconjunctival, sub-tendon, sub-tendon, peribulbar, peribulbar, posterior posterior juxta-scleral, juxta-scleral, and and retrobulbar retrobulbar spaces) spaces) without without performingperforming aa complicatedcomplicated surgery.surgery. Additionally,Additionally, suchsuch implantsimplants useuse aa shortershorter transscleraltransscleral routeroute thatthat allowsallows relativelyrelatively highhigh permeabilitypermeability ofof largerlarger drugsdrugs (up(up toto 7070kDa) kDa) [16[16,17,17].]. InIn addition,addition, wewe designeddesigned thesethese devicesdevices withwith aa singlesingle sidedsided permeablepermeable membranemembrane facingfacing thethe sclera,sclera, whichwhich increasedincreased thethe drugdrug deliverydelivery effiefficiencyciency by by reducing reducing drug drug elimination elimination by conjunctival by conjunctival clearance. clearance Although. theseAlthough minimally these invasiveminimally devices invasive allowed devices long-term allowed (18 long weeks-term [13 (18]) release weeks of [13] pre-loaded) release of drugs, pre-loaded they had drugs, to be replacedthey had onceto be thereplaced drug ranonce out. the Itdrug was ran also out. diffi Itcult was to also pre-determine difficult to pre the-determine exact time the for exact device time replacement. for device Thus,replacement a self-sustainable. Thus, a self way-sustainable of drug delivery way of drug is desirable. delivery is desirable. AA promisingpromising wayway toto achieveachieve self-sustainableself-sustainable drugdrug deliverydelivery isis toto replacereplace thethe drugsdrugs inin thethe devicedevice withwith geneticallygenetically modifiable modifiable cells cells that that cancan continuouslycontinuously secrete secrete trophic trophic factor factor proteins proteins [ 18[18]].. In In fact,fact, thisthis techniquetechnique hashas nownow gainedgained widewide popularitypopularity amongstamongst manymany researchresearch groups groups [ [55,19,19].]. Herein,Herein, wewe utilizedutilized aa retinalretinal pigmentpigment (RPE) cell cell line line (ARPE (ARPE-19;-19; [20] [20]).). The The RPE RPE cells cells play play an an important important role role in inthe the health health of of the the retina retina including including,, but but not not limited limited to, to, the the transport transport of of ions, nutrients, and water; water; absorptionabsorption ofof light; and protection against against photooxidation photooxidation [21 [21,22,22].]. RPE RPE cells cells can can also also be be modified, modified, in inprinciple, principle, to toproduce produce almost almost any any trophic trophic factors factors [18] [18],, which which makes makes it highly valuablevaluable forfor treatingtreating regenerativeregenerative diseases. diseases. Here,Here, wewe culturedcultured thethe ARPE-19ARPE-19 cellscells onon collagencollagen coatedcoated polystyrenepolystyrene (PS)(PS) sheetssheets andand transferredtransferred these cell cell-loaded-loaded sheets sheets to to a a3D 3D printed printed capsule capsule (Figure (Figure 1).1 ).Using Using the thedeveloped developed cell- cell-encapsulationencapsulation device, device, we wetested tested the the efficacy efficacy of ofthe the device device in in defending defending the the ARPE ARPE-19-19 cells fromfrom thethe body’sbody’s immuneimmune responseresponse (limiting(limiting didiffusionffusion of ofmolecules molecules biggerbigger thanthan 150150 kDa),kDa), whilewhile simultaneouslysimultaneously allowingallowing didiffusionffusion ofof oxygenoxygen andand nutrientsnutrients insideinside thethe device,device, andand releaserelease ofof BDNFBDNF toto thethe outsideoutside environmentenvironment (molecules (molecules smaller smaller than than 40 kDa). 40 kDa) Thus,. byThus, utilizing by utilizing advancement advancement in cell sheet in engineering cell sheet andengineering 3D printing, and we3D developedprinting, we a self-sustainable developed a self cell-encapsulation-sustainable cell- deviceencapsulation that has device the potential that has to the be usedpotential as a minimallyto be used invasiveas a minimally periocular invasive transport periocular for the transport treatment for of the retinal treatment diseases. of retinal diseases.

Figure 1. Overview of the cell-encapsulation device. (A) A 3D printed capsule with ARPE-19 cells enclosedFigure 1. inside Overview the device. of the ARPE-19cell-encapsulation cells were device. cultured (A in) A polystyrene 3D printed (PS) capsule sheets. with (B) ARPE Cross-section-19 cells ofenclosed device insi in A.de Thethe device. 3D printed ARPE capsule-19 cells with were semi-porous cultured in polystyrene membrane (PEGDM)(PS) sheets allowed. (B) Cross selective-section permeabilityof device in of A. brain-derived The 3D printed neurotrophic capsule with factor semi (BDNF;-porous 27 kDa), membrane O2, and (PEGDM) nutrients (16–180allowed Da) selective while protectingpermeability the ofcells brain from-derived the immune neurotrophic response factor of the ( bodyBDNF (i.e.,; 27 immunoglobulin kDa), O2, and nutrients (IgG; 150 (16 kDa)).–180 Da) while protecting the cells from the immune response of the body (i.e., immunoglobulin (IgG; 150 kDa)).

Micromachines 2020, 11, 436 3 of 12

2. Materials and Methods

2.1. Materials The following reagents and all other chemicals used in this study were commercially available and used without further purification: polystyrene (PS, Sigma-Aldrich, St. Louis, MO, USA); polyvinyl alcohol (PVA, Sigma-Aldrich), polyethylene glycol dimethacrylate (PEGDM, MW 750, Sigma-Aldrich); triethylene glycol dimethacrylate (TEGDM, MW 286.3, Sigma-Aldrich); FITC-dextran 40 (FD40, 40kDa, Sigma-Aldrich); FITC-dextran 150 (FD150, 150 kDa, Sigma-Aldrich); FIT7C-IgG (150 kDa, Sigma-Aldrich); Omnirad 819 (formerly Irgacure 819, IGM Resin B.V., Waalwijk, The Netherlands); 2-Isopropylthioxanthone (ITX; photosensitizer, Tokyo Chemical Industry Co., Ltd., Tokyo, Japan).

2.2. Resin Fabrication Two types of resins were fabricated: PEGDM and TEGDM. PEGDM (Mw 750) was mixed with an equal volume of water (1:1, v/v), while TEGDM (Mw 286.3) was used as it was. Then, 1% photo initiator (Irgacure 819, w/w), and 1% photosensitizer (ITX, w/w) were dissolved into each solution of PEGDM and TEGDM using a magnetic stirrer (4 h, 150 rpm) inside an aluminum wrapped plastic bottle (to protect the solution from light). The resins were filtered with a metallic mesh-type filter (coffee filter with 25 µm holes, Amazon, Tokyo, Japan) to separate aggregates of cured resins from previous printing.

2.3. 3D Printing of Cell Encapsulation Capsule The cell encapsulation capsule consisted of two parts: the cap and the reservoir. Both parts were designed in Solidworks 2018 (Dassault Systèmes, Vélizy-Villacoublay, France) and 3D printed (layer thickness: 50 um) (QiDi tech shadow 5.5 s, QIDI Technology Co., Ltd., Ruian, China) using either TEGDM (cap) or both TEGDM and PEGDM (reservoir) resins (Figure A1). To print the cap, each layer of TEGDM resin was exposed for 8 s (first layer was exposed for 30 s to allow adhesion to the build plate). However, to print the reservoir, a semi-autonomous approach was used. The first layer (exposure time: 30 s) was printed with TEGDM. Then, the printer was paused and the build plate was removed, followed by build plate cleaning with 95% ethanol. TEGDM resin was replaced with PEGDM, and second layer was printed (exposure time: 40 s). The printer was again paused, the build plate removed and washed with ethanol. Then, PEGDM resin was replaced with TEGDM, and the remaining layers were printed normally with TEGDM resin (exposure time: 8 s). After the parts were fully printed, they were washed with 95% ethanol for 1 min and post-cured in a UV box (50 W, 405 nm) for 5 min. Finally, both parts were soaked in 50% ethanol for 72 h to remove the photo initiator and the photosensitizer to improve biocompatibility of the device [23]. A detailed fabrication procedure is shown in Figure A1.

2.4. Cell Culture and Cell Counting The human retinal pigment epithelial cell line (ARPE-19) was purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA) and routinely sub-cultured (cell seeding density: 40,000 cell/cm2). They reached around 70% confluency, in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS, S 1400-500, Biowest, France) and 1% Antibiotic-Antimycotic (100X, Gibco, 15240062) at 37 ◦C in a humidified condition under 5% CO2. Once the cells were confluent, they were detached from the cell culture flask by adding 0.25% (w/v) trypsin and 0.1% (w/v) EDTA and incubating in a cell-culture incubator for 4 min. The detached cells were suspended in a fresh medium and centrifuged at 1500 rpm for 5 min, before removing the medium and resuspending the cells in fresh medium. To count the total number of cells, the first image was converted to RGB stack and the background was removed and converted to binary image. A 200 µm 200 µm image area was defined and image × Micromachines 2020, 11, 436 4 of 12 processing functions (fill hole, and convert to mask) were used to differentiate the cells from the background. Finally, particle analysis function was used to count the cells.

2.5. PS Sheet Fabrication The polystyrene (PS) sheet for cell culture was prepared by a combination of spin-coating and contact printing techniques, as previously described [24]. Briefly, polydimethylsiloxane (PDMS) stamps with convex columns were fabricated by conventional photolithography using SU-8 molds. PS was dissolved in dichloromethane to obtain a 400 mg/mL concentration of PS solution which was spin coated onto the PDMS stamp (4000 rpm, 10 s). Likewise, 10 mg/mL PVA solution was spin coated on a glass substrate (4000 rpm, 40 s). As shown in Figure A2A, the glass plate and the PDMS stamp were pressed against each other with the help of a clip, and baked (120 ◦C, 90 s). After baking, PS sheet was transformed onto the PVA coated glass plate. The PS patterned glass was then spin coated with 0.5 mg/mL collagen (4000 rpm, 40 s; Cell Matrix® type 1-A collagen, Nitta Co., Ltd.). The average diameter and thickness of the PS sheets were 5 mm and 150 nm, respectively.

2.6. Cell Encapsulation Inside the 3D Printed Capsule ARPE-19 cells were seeded onto the collagen coated PS sheet (diameter: 5 mm) with a cell density of 1.0 105 cells/mL. Cell were incubated for 2 h to allow adherence to the PS sheets. After that, × the glass plate containing the cell seeded PS sheets was gently washed with PBS- to remove excess and unattached cells. Then, cell loaded glass was dipped in a flask containing cell culture medium and incubated in the cell culture incubator. After 3–4 days, cells grew confluent on the PS sheet, and PVA got completely dissolved in the medium. The PS sheets were then removed from the glass substrate by first gently washing with cell culture medium and transferring it to the 3D printed capsule with the help of a tweezer. A total of 12 PS sheets were transferred into each 3D printed reservoir containing 690 µL medium. After transferring the cell containing PS sheets, and filling it with the cell culture medium, both the reservoir and cap were permanently bonded using PEGDM and UV curing for 30 s. The detailed fabrication procedure is shown in Figure A2B.

2.7. In-Vitro Diffusion Experiment with Semi-Porous Insert Device The semi-porous membrane was fabricated by curing the PEGDM resin, sandwiched between glass and a 100 µm thickness silicon mold, with a 365 nm UV lamp (Lightningcure LCS, Hamamatsu Photonics, Hamamatsu City, Japan) for 60 s at an intensity of 310 mW/cm2 (Figure A3A). The resulting membrane was glued using an acrylate based instant adhesive (Aron Alpha®, Tokyo, Japan) to the base of a 12-well insert (Falcon #353180) after removing the existing PET membrane from it. For the diffusion experiment, each insert was loaded with 800 µL FITC-conjugated dextran of either 40 kDa, 150 kDa, or FITC-IgG each of 100 µg/mL concentration, and the well was filled with 2 mL PBS- (Figure A3B). Here, dextran of different molecular weights was used as an evaluation of molecular transport and permeability. The 12-well plate was then put inside a humidified cell culture incubator at 37 ◦C for 7 days. An end point diffusion measurement was done by taking 200 µL sample from the well and measuring absorption intensity of the sample using a fluorescence plate reader (FluoroscanAscent, Thermo Fisher Scientific, Waltham, MA, USA). Based on a pre-calibrated standard curve, the amount of diffused reagent was determined.

2.8. Viability of ARPE-19 Cells Cultured on PS Sheets ARPE-19 cells cultured on PS sheets were stained using Calcein-AM and Propidium Iodide (PI; Cellstain® Double Staining Kit, Dojin Chemical Lab) at a concentration of 2 µmol/L and 4 µmol/L with PBS-, respectively, following the manufacturer’s staining protocol. Microscopic images were captured using a fluorescent microscope (Olympus, Tokyo, Japan) or a confocal microscope (LSM700, Carl Zeiss MicroImaging, Gina, Germany) on day 0, 9, 15, 22, and 30 and analyzed using the Image-J software (Fiji). The cell viability was measured by first converting the Calcein-AM stained (live cells) images to Micromachines 2020, 11, 436 5 of 12 Micromachines 2020, 11, x 5 of 12

images to 8-bit grayscale, then image thresholding function was used to eliminate noise. The 8-bit grayscale, then image thresholding function was used to eliminate noise. The percentage of live percentage of live cell coverage on PS sheets was used to obtain cellular viability. cell coverage on PS sheets was used to obtain cellular viability.

2.9. Quantification Quantification of BDNF Release from ARPE-19ARPE-19 Cells Two separateseparate experimentsexperiments were donedone toto calculatecalculate BDNFBDNF expressionexpression fromfrom ARPE-19ARPE-19 cells.cells. In the firstfirst group, BDNF release was measured by taking 100 µL of the medium sample (with replacement) ® from aa fullyfully confluentconfluent ARPE-19ARPE-19 monolayermonolayer culturedcultured inin aa 6-well6-well plateplate (Falcon(Falcon®, #353502),#353502), with 3 mLmL medium inside. In th thee second group,group, PS sheetssheets (with confluentconfluent ARPE-19ARPE-19 cells)cells) containingcontaining capsulescapsules werewere putput insideinside aa 6-well6-well plateplate containingcontaining 3 3 mL mL cell cell culture culture medium. medium. BDNF BDNF was was measured measured by by taking taking a 100a 100µL u mediumL medium sample sample from from each each well. well. To calculate To calculate the amount the amount of BDNF of BDNF secreted secreted by the by cells, the 100 cells,µL medium100 uL medium from each from sample each sample was measured was measured using the using ELISA the kitELISA (Rat kit BDNF (Rat ELISABDNF ELISA Kit, RayBio). Kit, RayBio). Based onBased a pre-calibrated on a pre-calibrated absorption absorption curve, the curve,mass of the secreted mass BDNF of secreted was determined BDNF was by multiplying determined the by concentrationmultiplying the obtained concentration by ELISA obtained to the amountby ELISA of to medium the amount sampled. of medium sampled.

3. Results and Discussion

3.1. In In-Vitro-Vitro DiDiffusionffusion Test The cell-encapsulationcell-encapsulation device should serve three functions: protection of cells from the body’sbody’s immuneimmune system; exchange of oxygen and nece necessaryssary nutrients; and release of BDNF. Molecular weight of nutrients range from 16–18016–180 Da [[25]25],, BDNFBDNF is 2727 kDakDa [[26]26],, andand immunoglobulinsimmunoglobulins are largerlarger thanthan 150150 kDa kDa [25] [25].. Thus, Thus, a a semi semi-porous-porous membrane should should be be able to allow diffusion of molecules bigger than 2727 kDa,kDa, butbut limit limit di diffusionffusion of of molecules molecules larger larger than than 150 150 kDa. kDa. As shownAs shown in Figure in Figure A3B, A3B, the in-vitro the in- divitroffusion diffusion test was test conducted was conducted in a transwell in a transwell insert byinsert replacing by replacing existing existing PET membranes PET membrane with 100s withµm thickness100 µm thickness PEGDM PE membranes.GDM membrane Figures. A3FigureC shows A3C theshows time the course time course of diffusion of diffusion of FD40 of overFD40 day over 7. Asday shown 7. As shown in Figure in 2Figure, the PEGDM 2, the PEGDM semi-porous semi- membraneporous membrane allowed allowed 43% of 40 43% kDa of molecules40 kDa molecules to pass through,to pass through while limiting, while limi bothting 150 both kDa 150 and kDa IgG and molecules IgG molecules to less thanto less 6% than diff usion.6% diffusion Although,. Although, FD150 andFD150 IgG and have IgG the have same the molecularsame molecular weight, weight, their ditheirffusion diffusion through through PEGDA PEGDA differ differ significantly. significantly. This couldThis could be due be to due the to di fftheerence differen in thece shapein the ofshape the molecules of the molecules (FD150 is(FD150 a string is likea string molecule, like molecule whereas, IgGwhereas is spherical IgG is spherical one). Likewise, one). Likewise, the diffusion the diffusion coefficient coefficient for FD40, for FD150, FD40, FD and150, IgG and were IgG 4.66 were 0.13,4.66 ± 0.072± 0.13, 0.0720.001, ± 0.001, and 0.63 and 0.630.01 ± 0.1001 ×14 10m-142/ ms,2 respectively,/s, respectively suggesting, suggesting that that the the PEGDM PEGDM membrane membrane is ± ± × − suitableis suitable for for use use in in construction construction of of the the cell cell encapsulation encapsulation device. device.

Figure 2. Graph of percentage diffusion of different FITC-conjugated dextran and immunoglobin G (IgG)Figure taken 2. Graph at day of 7 (npercentage= 3 for each diffusion group). of The different experiment FITC- wasconjugated conducted dextran in 12-well and immunoglobin transwell inserts G by(IgG) replacing taken at the day insert’s 7 (n = original3 for each PET group membrane). The experiment with a PEGDM was conducted membrane in (thickness: 12-well transwell 100 µm). inserts Error bar indicates standard deviation; ** and indicate p < 0.00005 and p < 0.00001 respectively. by replacing the insert’s original PET membrane†† with a PEGDM membrane (thickness: 100 μm). Error bar indicates standard deviation; ** and †† indicate p < 0.00005 and p < 0.00001 respectively.

Micromachines 2020, 11, 436 6 of 12

Micromachines 2020, 11, x 6 of 12 3.2. High Throughput Fabrication of Cell-Capsule 3.2. High Throughput Fabrication of Cell-Capsule 3D printing is gaining popularity in medical application [27]. Previously, we had utilized mold technique3D printing to fabricate is gaining drug popularity delivery devices in medical (DDS) application [14,15], but [27] it. wasPreviously, laborious. we Here,had utilized with the mold 3D techniqueprinting technique, to fabricate we drug were delivery able to shortendevices (DDS) device [14 prototyping,15], but it timewas (onlaborious. average, Here, mold with technique the 3D printingallowed technique 3–4 devices, we to were be made able into 1shorten h, whereas device 20–25 prototyping devices couldtime (on be printedaverage, in mold the sametechnique time allowedinterval using3–4 devices 3D printing) to be made and cutin 1 prototyping h, whereas time20–25 by devices almost could 4–5 times. be printed Thus, utilizingin the same the time new interval3D printing using technique, 3D printing) we and manufactured cut prototyping the cell-capsules time by almost in two4–5 times. parts: Thus, the cap utilizing and the the reservoir new 3D printing(detailed technique, dimensions we are manufactured given in Figure the A4 cell). As-capsules shown in in Figuretwo parts3, the: the cap cap was and 3D printed the reservoir with (detailnon-poroused dimension TEGDMs are material, given in while Figure the A4) reservoir. As shown was madein Figure by a3, semi-autonomousthe cap was 3D printed multi-material with non- porous3D printing TEGDM technique. material Although,, while the there reservoir were was few multi-materialmade by a semi 3D-autonomous printers on multi the market,-material these 3D printingprinters weretechni designedque. Although, for industrial there were applications few multi only-material and are 3D quite printers expensive. on the market, Therefore, these we printers utilized werea semi-manual designed for technique industrial whereby applications we printed only aand part are up quite to the expensive. desired height, Therefore then, we paused utilized the a printer semi- manualand replaced technique the material whereby with we another printed one a part (Figure up toA1 the). This desired process height was, then repeated paused until the we printer achieved and a replaceddesired multi-material the material with capsule. another The majorone (Figure advantage A1). ofThis 3D process printing was is that repeated it allows until fast we prototyping achieved ofa desiredcomplex multi 3D geometry-material thatcapsule. is not The possible major with advantage conventional of 3D printing manufacturing is that it techniques allows fast like prototyping milling or oflaser complex cutting. 3D Although geometry we that used is not a simple possible cylindrical with conventional box design manufacturing for in-vitro tests, techniques it should like be possiblemilling orto 3Dlaser print cutting. a more Although realistic we periocular used a simple transplant cylindrical device withbox design a curved for platein-vitro geometry tests, it that should fits thebe possiblecurvature to of 3D the print eye. a more realistic periocular transplant device with a curved plate geometry that fits the curvature of the eye.

Figure 3. 3D printedprinted cellcell encapsulation encapsulation capsule. capsule. (A ()A 3D) 3D exploded exploded view view of the of CADthe CAD model. model. (B) Top (B) viewTop viewof the of actual the fabricated actual fabricated device (C device) Magnified (C) Magnified view of the view reservoir of the showing reservoir multi-material showing multi printing-material of a printingsemi-porous of a semi PEGDM-porous membrane PEGDM and membrane non-porous and TEGDM non-porous outer TEGDM covering. outer covering. 3.3. Cell Viability of ARPE-19 Cells Cultured in Polystyrene (PS) Sheet 3.3. Cell Viability of ARPE-19 Cells Cultured in Polystyrene (PS) Sheet Since a self-sustainable cell-encapsulation device needs to function for several months to years, Since a self-sustainable cell-encapsulation device needs to function for several months to years, viable cells at the optimal state are paramount. Here, we cultured ARPE-19 cell on a collagen coated viable cells at the optimal state are paramount. Here, we cultured ARPE-19 cell on a collagen coated PS sheet (Figure4A). Cell sheet technology has been widely utilized for cellular transplant in tissue PS sheet (Figure 4A). Cell sheet technology has been widely utilized for cellular transplant in tissue engineering and regenerative medicine applications [24,28]. Utilizing this technique, we successfully engineering and regenerative medicine applications [24,28]. Utilizing this technique, we successfully cultured ARPE-19 cells on collagen coated PS sheets. As shown in Figure4B, cells cultured on PS cultured ARPE-19 cells on collagen coated PS sheets. As shown in Figure 4B, cells cultured on PS sheets maintained more than 75% viability for at least 30 days. ARPE-19 cells form monolayers with sheets maintained more than 75% viability for at least 30 days. ARPE-19 cells form monolayers with a polygonal shape [20]. As can be seen in Figure4C,D, ARPE-19 grew confluent and maintained a a polygonal shape [20]. As can be seen in Figure 4C,D, ARPE-19 grew confluent and maintained a polygonal shape on day 30. Likewise, as shown in Figure5A, ARPE-19 containing PS sheets were polygonal shape on day 30. Likewise, as shown in Figure 5A, ARPE-19 containing PS sheets were enclosed inside the cell encapsulation capsule. PS sheets taken out from the capsule on day 3 also show enclosed inside the cell encapsulation capsule. PS sheets taken out from the capsule on day 3 also good cell viability (Figure4B). A total of 12 PS sheets (5 mm diameter each) was enclosed inside a show good cell viability (Figure 4B). A total of 12 PS sheets (5 mm diameter each) was enclosed inside a capsule diameter (20 mm) × height (4.15 mm), however the sheet density inside the capsule can be increased by utilizing micro sheets [28].

Micromachines 2020, 11, 436 7 of 12

Micromachines 2020, 11, x 7 of 12 capsule diameter (20 mm) height (4.15 mm), however the sheet density inside the capsule can be Micromachines 2020, 11, x × 7 of 12 increased by utilizing micro sheets [28].

Figure 4. Cellular viability of ARPE-19 cells in collagen coated PS sheets. (A) Cells were cultured on Figure 4. Cellular viability of ARPE-19 cells in collagen coated PS sheets. (A) Cells were cultured on collagen coated 5 mm diameter PS sheets. (B) ARPE-19 cells maintain over 75% cell viability for at collagenFigure 4. coated Cellular 5 mm viability diameter of ARPE PS sheets.-19 cells (B) in ARPE-19 collagen cells coated maintain PS she overets. 75%(A) cellCells viability were cultured for at least on least 30 days on the PS sheet (n = 3). Morphology of confluent monolayer cells are similar for both (C) 30collagen days on coated the PS 5 sheetmm diameter (n = 3). Morphology PS sheets. ( ofB) confluentARPE-19 monolayercells maintain cells over are similar 75% cell for viability both (C) for day at 0 day 0 and (D) day 30. andleast ( D30) days day 30.on the PS sheet (n = 3). Morphology of confluent monolayer cells are similar for both (C) day 0 and (D) day 30.

Figure 5. Cellular viability of ARPE-19 cells inside cell encapsulation capsule. (A) Conceptual image of Figurethe cell 5. cultured Cellular PS viability sheets encapsulationof ARPE-19 cells process. inside ( Bcell) Image encapsulation of ARPE-cells capsule. in 5 mm(A) Conceptual PS sheets on image day 3. of the cell cultured PS sheets encapsulation process. (B) Image of ARPE-cells in 5 mm PS sheets on 3.4. BDNFFigure Release5. Cellular from viability the Cell-Encapsulated of ARPE-19 cells Device inside cell encapsulation capsule. (A) Conceptual image day 3. of the cell cultured PS sheets encapsulation process. (B) Image of ARPE-cells in 5 mm PS sheets on Two tests were conducted to evaluate BDNF secretion from ARPE-19 cells. The first test (control day 3. 3.4.group) BDNF consisted Release from of a the confluent Cell-Encapsulated monolayer Device of ARPE-19 cells grown in a 6-well culture plate, while the second one (experimental group) contained confluent cells grown on 12 PS sheets that were 3.4. TwoBDNF tests Release were from conducted the Cell- toEncapsulated evaluate BDNF Device secretion from ARPE-19 cells. The first test (control group)embedded consisted inside of the a confluent capsule. Figuremonolayer6 shows of ARPE cumulative-19 cells BDNF grown release in a 6- inwell both culture groups. plate, As while expected, the Two tests were conducted to evaluate BDNF secretion from ARPE-19 cells. The first test (control secondBDNF secretionone (experimental increased monotonicallygroup) contained in theconfluent control groupcells grown (6-well on plate) 12 PS and she theets experimental that were group) consisted of a confluent monolayer of ARPE-19 cells grown in a 6-well culture plate, while the embeddedgroup (capsule). inside the In thecapsule. experimental Figure 6 groupshows incumulative particular, BDNF BDNF release secretion in both was groups. observed As for expected, 16 days. second one (experimental group) contained confluent cells grown on 12 PS sheets that were BDNFHowever, secretion total BDNFincreased release monotonically from the capsule in the wascontrol around group 3.5 (6 times-well (normalized plate) and th toe totalexperimental cell count embedded inside the capsule. Figure 6 shows cumulative BDNF release in both groups. As expected, groupand percentage (capsule). In of BDNFthe experimental diffusion through group in the particular, capsule) BDNF lower secretion than the onewas measured observed infor the 16 6-welldays. BDNF secretion increased monotonically in the control group (6-well plate) and the experimental However,culture plate total (in BDNF reference release to datafrom from the capsule day 7). Onewas possiblearound 3 reason.5 times that (normalized a low BDNF to wastotal observed cell count in group (capsule). In the experimental group in particular, BDNF secretion was observed for 16 days. andthe percentage experimental of sampleBDNF diffusion compared through to the control the capsule) could belower because than BDNF the one gradient measured inside in thethe capsule6-well However, total BDNF release from the capsule was around 3.5 times (normalized to total cell count cultureand outside plate the(in reference capsule was to data much from smaller day 7). initially. One possible In addition, reason for that actual a low therapeutic BDNF was eff ectobserved on retinal in and percentage of BDNF diffusion through the capsule) lower than the one measured in the 6-well thedegenerative experimental diseases, sample administrationcompared to the of control several could ng/mL be because or more BDNF is necessary gradient [2 inside,3]. Although, the capsule we culture plate (in reference to data from day 7). One possible reason that a low BDNF was observed in and outside the capsule was much smaller initially. In addition, for actual therapeutic effect on retinal the experimental sample compared to the control could be because BDNF gradient inside the capsule degenerative diseases, administration of several ng/mL or more is necessary [2,3]. Although, we and outside the capsule was much smaller initially. In addition, for actual therapeutic effect on retinal utilized a non-engineered ARPE-19 cell line as a proof-of-concept study, it is possible to increase degenerative diseases, administration of several ng/mL or more is necessary [2,3]. Although, we

utilized a non-engineered ARPE-19 cell line as a proof-of-concept study, it is possible to increase

Micromachines 2020, 11, 436 8 of 12

Micromachinesutilized a non-engineered 2020, 11, x ARPE-19 cell line as a proof-of-concept study, it is possible to increase8 BDNF of 12 secretion by genetically engineering ARPE-19 cells. In fact, such transgenic technology already exists BDNFand several secretion groups by genetically have reported engineering secretion ARPE of several-19 cells. neurotrophic In fact, such factors, transgenic including technology BDNF already in the existsorder and of ng several/mL [29 groups–32]. Utilizinghave reported such secretion genetically of several engineered neurotrophic ARPE cells, factors it should, including be possible BDNF in to theimprove order theof ng performance/mL [29–32] of. Utilizing the device such further. genetically engineered ARPE cells, it should be possible to improve the performance of the device further.

Figure 6. Total amount of BDNF collected from ARPE-19 cultured in (A) 6-well plate and (B) Figurecell-encapsulated 6. Total amount capsule. of ImageBDNF insertscollected show from illustration ARPE-19 of cultured the experiment in (A) 6 conducted.-well plate Theand error (B) cell bars- encapsulatedindicate standard capsule. deviation Image (n inserts= 3 samples show forillustration each group). of the experiment conducted. The error bars indicate standard deviation (n = 3 samples for each group). 4. Conclusions and Future Work 4. Conclusions and Future Work We built a cell encapsulation device by 3D printing a semi-porous capsule, with ARPE-19 cells grownWe on built collagen a cell coatedencapsulation PS sheets device enclosed by 3D inside. printing Prior a to semi 3D- printingporous capsule the semi-porous, with ARPE capsule,-19 cells we grownconducted on collage diffusionn coated experiments PS she andets enclosed observed inside. that the Prior photocurable to 3D printing PEGDM the membrane semi-porous was capsule, capable weof limitingconducted diff usiondiffusion of large experiments molecules and (150 observed kDa) to that protect the thephotocurable cells from thePEGDM body’s membrane immune factor was capablemolecules, of limiting while simultaneously diffusion of large allowing molecules passage (150 kDa) of smaller to protect molecules the cells (40 from kDa) the likebody’s oxygen immune and factornutrients molecules, that are essentialwhile simultaneously for cell survivability, allowing and passage BDNF of molecules smaller thatmolecules are needed (40 kDa) for the like treatment oxygen andof retinal nutrients diseases. that are Further, essential ARPE-19 for cell cells survivability cultured on, collagenand BDNF coated molecules PS sheets that were are ableneeded to maintain for the treatment75% confluency of retina forl diseases. at least 30Further, daysoutside ARPE-19 the cells device, cultured and on were collagen able to coated secrete PS BDNF sheets for were at leastable to16 maintain days while 75% inside confluency the device. for at The least proposed 30 days cell-encapsulation outside the device, device and were has the able potential to secrete to beBDNF used for as ata self-sustainableleast 16 days while way inside to deliver the BDNFdevice. to The the proposed retina through cell-encapsulati minimallyon invasive device periocular has the potential transplant. to beIn used the future, as a self we-sustainable plan to increase way to BDNF deliver release BDNF further to the by retina utilizing through genetically minimally modified invasive ARPE-19 periocular cells transplant.and testing In the the therapeutic future, we potential plan to increase on an animal BDNF model. release further by utilizing genetically modified ARPE-19 cells and testing the therapeutic potential on an animal model. Author Contributions: Conceptualization, H.K. (Hirokazu Kaji), H.K. (Hideto Kojima), and T.A.; methodology Authorand fabrication Contributions: of the devices, Conceptualization, H.K. (Hideto H.K Kojima)aji, H. andKojima B.R.;, dataand analysis, T.A.; methodology H.K. (Hideto and Kojima), fabrication B.R., L.-J.C.,of the devices,N.N., T.A., H.Kojima and H.K. and B.R.; (Hirokazu data analysis, Kaji); writing—original H.Kojima, B.R., L. draft-J.C., N.N., preparation, T.A., and H.K. H.Kaji (Hideto; writing Kojima) – original and draft B.R.; preparation,writing—reviewing H.Kojima and and editing, B.R.; H.K.writing (Hideto – reviewing Kojima), and B.R., editing, L.-J.C., H.K. N.N., (Hideto T.A., and Kojima) H.K., (Hirokazu B.R, L.-J.C. Kaji),, N.N., funding T.A., acquisition, H.K. (Hirokazu Kaji). All authors have read and agreed to the published version of the manuscript. and H.Kaji, funding acquisition, H.Kaji. All authors have read and agreed to the published version of the manuscript.Funding: This work was funded by a Grant-in-Aid for Scientific Research (B) (17H02752) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan. Funding: This work was funded by a Grant-in-Aid for Scientific Research (B) (17H02752) from the Ministry of Conflicts of Interest: The authors declare no conflict of interest. Education, Culture, Sports, Science and Technology (MEXT), Japan.

Conflicts of Interest: The authors declare no conflict of interest.

Micromachines 2020, 11, 436 9 of 12 Micromachines 2020, 11, x 9 of 12 Micromachines 2020, 11, x 9 of 12 AppendixAppendix AA Appendix A

Figure A1. 3D printing cell-encapsulation capsule. The cap was printed with TEGDM, whereas the Figure A1. 3D printing cell-encapsulation capsule. The cap was printed with TEGDM, whereas the Figurereservoir A1. was 3D printedprinting with cell- encapsulationboth TEGDM capsuleand PEGDM.. The cap After was 3D printed printing, with both TEGDM parts ,were whereas washed the reservoir was printed with both TEGDM and PEGDM. After 3D printing, both parts were washed with reservoirwith 95% was ethanol printed for 1with min ,both followed TEGDM by post and- curingPEGDM. in UVAfter box 3D (50 printing, W, 405 nm)both for parts 5 min were. Finally, washed 3D 95% ethanol for 1 min, followed by post-curing in UV box (50 W, 405 nm) for 5 min. Finally, 3D printed withprinted 95% parts ethanol were for soaked 1 min ,in followed 50% ethanol by post for- curing72 h to in remove UV box photo (50 W, initiator 405 nm) and for photosensitizer 5 min. Finally, and3D parts were soaked in 50% ethanol for 72 h to remove photo initiator and photosensitizer and increase increase biocompatibility of the capsule. biocompatibilityprinted parts were of soaked the capsule. in 50% ethanol for 72 h to remove photo initiator and photosensitizer and increase biocompatibility of the capsule.

Figure A2. Fabrication of PS sheet, cell culturing, and cell encapsulation. (A) PS sheets were patterned inFigure a glass A2. substate Fabrication coated of PS with sheet, PVA. cell Collagen culturing, was and spin cell coatedencapsulation. on the PS (A sheet) PS sheets to culture were ARPE-19patterned cells.Figurein a glass (B A2.) ARPE-19 substateFabrication cells coated of cultured PS with sheet, PVA. on cell the Collagenculturing, PS sheet was couldand spincell be encapsulation. easilycoated removed on the (PSA once) sheetPS thesheets to PVA culture were substrate patterned ARPE was-19 dissolved.incells. a glass (B) ARPEsubstate Cells- were19 cellscoated then cultured encapsulatedwith PVA. on the Collagen PS inside sheet the was could capsule spin be coatedeasily by bonding removed on the the PS caponce sheet and the to thePVA culture reservoir substrate ARPE with was-19 a UVcells.dissolved. curable (B) ARPE Cells PEGDM-19 were cells resin. then cultured encapsulated on the PS inside sheet thecould capsule be easily by bonding removed the once cap the and PVA the substratereservoir withwas dissolved.a UV curable Cells PEGDM were then resin encapsulated. inside the capsule by bonding the cap and the reservoir with a UV curable PEGDM resin.

Micromachines 2020, 11, x 10 of 12

MicromachinesMicromachines2020 2020,,11 11,, 436x 1010 of 1212

Figure A3. Fabrication of PEGDM membrane and in-vitro diffusion test. (A) PEGDM resin sandwiched between glass and a 100 µm thickness silicon-sheet with a round cut was cured using 365 Figure A3. Fabrication of PEGDM membrane and in-vitro diffusion test. (A) PEGDM resin sandwiched nm UV lamp for 60 s. (B) The 12-well transwell insert membrane was replaced with a PEGDM between glass and a 100 µm thickness silicon-sheet with a round cut was cured using 365 nm UV membraneFigure A3. for Fabrication diffusion test of . PEGDM (C) Percentage membrane diffusion and in of- vitro FD40 diffusion (n = 3). The test. error (A) bar PEGDM represent resins lamp for 60 s. (B) The 12-well transwell insert membrane was replaced with a PEGDM membrane for standardsandwiched deviation. between glass and a 100 µm thickness silicon-sheet with a round cut was cured using 365 dinmffusion UV lamp test. (C for) Percentage 60 s. (B) The diff usion12-well of FD40 transwell (n = 3). insert The m errorembrane bar represents was replaced standard with deviation. a PEGDM membrane for diffusion test. (C) Percentage diffusion of FD40 (n = 3). The error bar represents standard deviation.

Figure A4. Detail dimension of the cell encapsulation capsule showing the isometric, top, and side view of (A) the cap and (B) thethe reservoir. DimensionDimension labelinglabeling isis inin mm.mm.

Figure A4. Detail dimension of the cell encapsulation capsule showing the isometric, top, and side view of (A) the cap and (B) the reservoir. Dimension labeling is in mm.

Micromachines 2020, 11, 436 11 of 12

References

1. Tekin, M.I.; Sekeroglu, M.A.; Demirtas, C.; Tekin, K.; Doguizi, S.; Bayraktar, S.; Yilmazbas, P. Brain-derived neurotrophic factor in patients with age-related macular degeneration and its correlation with retinal layer thicknesses. Investig. Ophthalmol. Vis. Sci. 2018, 59, 2833–2840. [CrossRef][PubMed] 2. Meyer-Franke, A.; Kaplan, M.R.; Pfieger, F.W.; Barres, B.A. Characterization of the signaling interactions that promote the survival and growth of developing retinal ganglion cells in culture. Neuron 1995, 15, 805–819. [CrossRef] 3. Harada, C.; Guo, X.; Namekata, K.; Kimura, A.; Nakamura, K.; Tanaka, K.; Parada, L.F.; Harada, T. Glia-and neuron-specific functions of TrkB signalling during retinal degeneration and regeneration. Nat. Commun. 2011, 2.[CrossRef][PubMed] 4. Frade, J.M.; Bovolenta, P.; Martínez-Morales, J.R.; Arribas, A.; Barbas, J.A.; Rodríguez-Tébar, A. Control of early cell death by BDNF in the chick retina. Development 1997, 124, 3313–3320. [PubMed] 5. Orive, G.; Santos-Vizcaino, E.; Pedraz, J.L.; Hernandez, R.M.; Ramirez, J.E.V.; Dolatshahi-Pirouz, A.; Khademhosseini, A.; Peppas, N.A.; Emerich, D.F. 3D cell-laden polymers to release bioactive products in the eye. Prog. Retin. Eye Res. 2019, 68, 67–82. [CrossRef][PubMed] 6. Kaji, H.; Nagai, N.; Nishizawa, M.; Abe, T. Drug delivery devices for retinal diseases. Adv. Drug Deliv. Rev. 2018, 128, 148–157. [CrossRef] 7. Poduslo, J.F.; Curran, G.L. MOLECULAR Permeability at the blood-brain and blood-nerve barriers of the neurotrophic factors: NGF, CNTF, NT-3, BDNF. Mol. Brain Res. 1996, 36, 280–286. [CrossRef] 8. Khalin, I.; Alyautdin, R.; Kocherga, G.; Bakar, M.A. Targeted delivery of brain-derived neurotrophic factor for the treatment of blindness and deafness. Int. J. Nanomed. 2015, 10, 3245–3267. [CrossRef] 9. Duvvuri, S.; Majumdar, S.; Mitra, A.K. Drug delivery to the retina: Challenges and opportunities. Expert Opin. Biol. Ther. 2003, 3, 45–56. [CrossRef] 10. Kim, Y.C.; Chiang, B.; Wu, X.; Prausnitz, M.R. Ocular delivery of macromolecules. J. Control Release 2014, 190, 172–181. [CrossRef][PubMed] 11. Touahri, Y.; Dixit, R.; Kofoed, R.H.; Miloska, K.; Park, E.; Raeisossadati, R.; Markham-Coultes, K.; David, L.A.; Rijal, H.; Zhao, J.; et al. Focused ultrasound as a novel strategy for noninvasive gene delivery to retinal Müller glia. Theranostics 2020, 10, 2982–2999. [CrossRef] 12. Nagai, N.; Saijo, S.; Song, Y.; Kaji, H.; Abe, T. A drug refillable device for transscleral sustained drug delivery to the retina. Eur. J. Pharm. Biopharm. 2019, 136, 184–191. [CrossRef][PubMed] 13. Nagai, N.; Nezhad, Z.K.; Daigaku, R.; Saijo, S.; Song, Y.; Terata, K.; Hoshi, A.; Nishizawa, M.; Nakazawa, T.; Kaji, H.; et al. Transscleral sustained ranibizumab delivery using an episcleral implantable device: Suppression of laser-induced choroidal in rats. Int. J. Pharm. 2019, 567, 118458. [CrossRef][PubMed] 14. Kawashima, T.; Nagai, N.; Kaji, H.; Kumasaka, N.; Onami, H.; Ishikawa, Y.; Osumi, N.; Nishizawa, M.; Abe, T. A scalable controlled-release device for transscleral drug delivery to the retina. Biomaterials 2011, 32, 1950–1956. [CrossRef][PubMed] 15. Nagai, N.; Kaji, H.; Onami, H.; Ishikawa, Y.; Nishizawa, M.; Osumi, N.; Nakazawa, T.; Abe, T. A polymeric device for controlled transscleral multi-drug delivery to the posterior segment of the eye. Acta Biomater. 2014, 10, 680–687. [CrossRef][PubMed] 16. Raghava, S.; Hammond, M.; Kompella, U.B. Periocular routes for retinal drug delivery. Expert Opin. Drug Deliv. 2004, 1, 99–114. [CrossRef][PubMed] 17. Ranta, V.P.; Urtti, A. Transscleral drug delivery to the posterior eye: Prospects of pharmacokinetic modeling. Adv. Drug Deliv. Rev. 2006, 58, 1164–1181. [CrossRef] 18. Emerich, D.F.; Orive, G.; Thanos, C.; Tornoe, J.; Wahlberg, L.U. Encapsulated cell therapy for neurodegenerative diseases: From promise to product. Adv. Drug Deliv. Rev. 2014, 67–68, 131–141. [CrossRef] 19. Lathuilière, A.; Mach, N.; Schneider, B.L. Encapsulated cellular implants for recombinant protein delivery and therapeutic modulation of the immune system. Int. J. Mol. Sci. 2015, 16, 10578–10600. [CrossRef] 20. Dunn, K.C.; Aotaki-Keen, A.E.; Putkey, F.R.; Hjelmeland, L.M. ARPE-19, a human retinal pigment epithelial cell line with differentiated properties. Exp. Eye Res. 1996, 62, 155–170. [CrossRef] Micromachines 2020, 11, 436 12 of 12

21. Strauss, O. The retinal pigment epithelium in visual function. Physiol. Rev. 2005, 85, 845–881. [CrossRef] [PubMed] 22. Chen, L.-J.; Ito, S.; Kai, H.; Nagamine, K.; Nagai, N.; Nishizawa, M.; Abe, T.; Kaji, H. Microfluidic co-cultures of retinal pigment epithelial cells and vascular endothelial cells to investigate choroidal angiogenesis. Sci. Rep. 2017, 7, 3538. [CrossRef][PubMed] 23. Urrios, A.; Parra-Cabrera, C.; Bhattacharjee, N.; Gonzalez-Suarez, A.M.; Rigat-Brugarolas, L.G.; Nallapatti, U.; Samitier, J.; DeForest, C.A.; Posas, F.; Garcia-Cordero, J.L.; et al. 3D-printing of transparent bio-microfluidic devices in PEG-DA. Lab Chip 2016, 16, 2287–2294. [CrossRef][PubMed] 24. Suzuki, J.; Nagai, N.; Nishizawa, M.; Abe, T.; Kaji, H. Electrochemical manipulation of cell populations supported by biodegradable polymeric nanosheets for cell transplantation therapy. Biomater. Sci. 2017, 5, 216–222. [CrossRef] 25. Chang, T.M.S.; Prakash, S. Therapeutic uses of microencapsulated genetically engineered cells. Mol. Med. Today 1998, 4, 221–227. [CrossRef] 26. Rosenfeld, R.D.; Zeni, L.; Haniu, N.; Talvenheimo, J.; Radka, S.F.; Bennett, L.; Miller, J.A.; Welcher, A.A. Purification and identification of brain-derived neurotrophic factor from human serum. Protein Expr. Purif. 1995, 6, 465–471. [CrossRef] 27. Aimar, A.; Palermo, A.; Innocenti, B. The Role of 3D Printing in Medical Applications: A State of the Art. J. Healthc. Eng. 2019, 2019, 5340616. [CrossRef] 28. Li, M.; Ma, J.; Gao, Y.; Yang, L. Cell sheet technology: A promising strategy in regenerative medicine. Cytotherapy 2019, 21, 3–16. [CrossRef] 29. Scheper, V.; Schwieger, J.; Hamm, A.; Lenarz, T.; Hoffmann, A. BDNF-overexpressing human mesenchymal stem cells mediate increased neuronal protection in vitro. J. Neurosci. Res. 2019, 97, 1414–1429. [CrossRef] 30. Abe, T.; Tokita-Ishikawa, Y.; Onami, H.; Katsukura, Y.; Kaji, H.; Nishizawa, M.; Nagai, N. Intrascleral transplantation of a collagen sheet with cultured brain-derived neurotrophic factor expressing cells partially rescues the retina from damage due to acute high intraocular pressure. Adv. Exp. Med. Biol. 2014.[CrossRef] 31. Kontturi, L.S.; Collin, E.C.; Murtomäki, L.; Pandit, A.S.; Yliperttula, M.; Urtti, A. Encapsulated cells for long-term secretion of soluble VEGF receptor 1: Material optimization and simulation of ocular drug response. Eur. J. Pharm. Biopharm. 2015, 95, 387–397. [CrossRef][PubMed] 32. Kauper, K.; McGovern, C.; Sherman, S.; Heatherton, P.; Rapoza, R.; Stabila, P.; Dean, B.; Lee, A.; Borges, S.; Bouchard, B.; et al. Two-year intraocular delivery of ciliary neurotrophic factor by encapsulated cell technology implants in patients with chronic retinal degenerative diseases. Investig. Ophthalmol. Vis. Sci. 2012, 53, 7484–7491. [CrossRef][PubMed]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).