Edited by: Dimitrios I. Zeugolis, National University of Ireland Galway, Ireland
Reviewed by: Christine A. Curcio, University of Alabama at Birmingham, United States; Thomas Ach, University Hospital Würzburg, Germany
This article was submitted to Tissue Engineering and Regenerative Medicine, a section of the journal Frontiers in Bioengineering and Biotechnology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Atrophic age-related macular degeneration (AMD) is the most common form of AMD accounting for 90% of patients. During atrophic AMD the waste/exchange pathway between the blood supply (choroid) and the retinal pigment epithelium (RPE) is compromised. This results in atrophy and death of the RPE cells and subsequently the photoreceptors leading to central blindness. Although the mechanisms behind AMD are unknown, the growth of fatty deposits known as drusen, have been shown to play a role in the disease. There is currently no treatment or cure for atrophic AMD. Much research focuses on developing a synthetic substrate in order to transplant healthy cells to the native Bruch’s membrane (BM), however, the diseased native BM and related structures still leave potential for transplanted cells to succumb to disease. In this proof-of-concept work we electrospun poly(ethylene terephthalate) (PET) to fabricate a nanofibrous cytocompatible synthetic BM. The apical surface of the membrane was cultured with ARPE-19 cells and the underside was decorated with poly(lactic acid-co-glycolic acid) (PLGA) or poly(glycolic acid) (PGA) degradable nanoparticles by electrospraying. The membrane exhibited hydrophilicity, high tensile strength and structurally resembled the native BM. ARPE-19 cells were able to form a monolayer on the surface of the membrane and no cell invasion into the membrane was seen. The presence of both PLGA and PGA nanoparticles increased ARPE-19 cell metabolism but had no effect on cell viability. There was a decrease in pH of ARPE-19 cell culture media 7 days following culturing with the PLGA nanoparticles but this change was eliminated by 2 weeks; PGA nanoparticles had no effect on cell culture media pH. The fluorescent dye FITC was encapsulated into nanoparticles and showed sustained release from PLGA nanoparticles for 2 weeks and PGA nanoparticles for 1 day. Future work will focus on encapsulating biologically active moieties to target drusen. This could allow this novel bioactive substrate to be a potential treatment for atrophic AMD that would function two-fold: deliver the required monolayer of healthy RPE cells to the macula on a synthetic BM and remove diseased structures within the retina, restoring the waste/exchange pathway and preventing vision loss.
Age-related macular degeneration (AMD) is a progressive disease of the retina that is the leading form of blindness in developed countries. It is a form of central blindness that mainly affects people over the age of 50 years. Although the pathology of AMD is currently unknown, age is considered to be the main contributing factor to the manifestation of this disease (
Much research focuses on developing a synthetic cell transplantation substrate in order to transplant healthy RPE cells onto BM as a treatment for nonexudative AMD (
Recent research has explored using an anti-inflammatory, antiatherogenic peptide L-4F to reduce the accumulation of fatty deposits on BM of
There is, therefore, scope to provide a two-pronged approach for a potential treatment for nonexudative AMD. That is, a persistent bioactive substrate that would function two-fold; provide a permanent basement layer for transplantation of healthy RPE cells to the area required, while removing diseased structures (drusen) on the BM (
Poly(ethylene terephthalate) (PET) pellets (04301, Polysciences Inc.) were dissolved in neat 1,1,1-3,3,3-hexafluoroisopropanol (HFIP) (Apollo Scientific Ltd.) at a concentration of 17.5% (w/v). The fibres were collected for 15 min under laboratory conditions on a grounded plate covered with aluminium foil at a working distance of 15 cm, flow rate of 2 ml/h and a voltage of 25 kV.
Poly(lactic acid-co-glycolic acid) 50/50 (PLGA) (26269, Polysciences Inc.) was dissolved in neat chloroform (Merck) at a concentration of 2% (w/v). Poly(glycolic acid) (06525, Polysciences Inc.) was dissolved in neat HFIP at a concentration of 1% (w/v) and left to stir overnight at room temperature. To encapsulate FITC into the nanoparticles, prior to electrospraying 2 mM of FITC was added to the polymer solutions and mixed thoroughly. The homogeneous polymer solution was introduced into a 10 mL plastic syringe and a blunt-tip needle attached. Any air bubbles were removed and the filled syringe was fixed in a mechanical syringe pump. All work was carried out at ambient conditions in a fume-hood. To assess initial nanoparticle formation, nanoparticles were sprayed into a glass dish filled with 0.1% (v/v) isopropanol (Merck) in dH2O placed on a magnetic stirrer, connected to a grounding plate and left to spray for 1 h. Nanoparticles were then decorated on the basal surface of the electrospun membrane using the optimal electrospraying parameters. Working parameters for PGA nanoparticles were 25 kV, 25 cm working distance, 2 ml/h flow rate; and for PLGA nanoparticles; 11 kV, 13 cm working distance, 0.5 ml flow rate (
Diameter measurements were undertaken to determine fibre/nanoparticle size and homogeneity between fabrication batches. The diameter of 50 fibres and nanoparticles from three different images of the electrospun fibres and electrosprayed nanoparticles were measured using ImageJ (
Scanning electron microscopy (SEM) was undertaken to characterise the morphology of the fibres and nanoparticles. Electrospun membrane or 10 μl of electrosprayed nanoparticles suspended in solution was placed on a carbon tab (TAAB) mounted on an aluminium stub (TAAB). The nanoparticles were surrounded by a layer of silver dag (Merck) and were left overnight in a desiccator for the solution to evaporate. Membrane or nanoparticles were gold sputter coated (Quorum) and imaged using SEM (Quanta FEG250 ESEM) with EHT of 5 kV (
To measure the wettability and identify if membranes were hydrophilic or hydrophobic, WCA measurements were carried out. Electrospun membranes were cut into 3 cm × 1 cm rectangles. Samples were either untreated, UV treated (1 h), placed in ethanol, or placed in cell culture medium and left to dry before being measured for changes in wettability using sessile-drop goniometry on the DSA 100 (Kruss-Scientific) (
Mechanical properties of the electrospun membranes were determined using tensile testing. Quantitative tensile testing of the electrospun membrane was undertaken using UniVert tensile tester (CellScale) equipped with a 10 N load cell at a displacement rate of 12 mm/min. The membranes were cut into dog-bone shaped strips 2 cm in length by 0.5 cm in width and tested until failure or until the tensile tester had reached maximum distance (
Fourier-transform infrared spectroscopy measurements were undertaken to determine changes in surface chemistry following membrane surface treatment. Electrospun membranes were cut into 3 cm × 1 cm rectangles. Samples were either untreated, UV treated (1 h), placed in ethanol, or placed in cell culture medium and left to dry before being measured using Vertex 70 Fourier Transform Infrared Spectrometer (Vertex) (
Electrospun membrane was cut into 1.5 cm2 squares and placed in Scaffdex (Merck). ARPE-19 cells (ATCC-LGC, CRL-2302 passage 28–30) were seeded at a density of 50,000 cells/sample and incubated at 37°C, 5% CO2, 98–99 % humidity and grown for 1 month or 3 months. Media consisted of DMEM:F12 (Merck) containing 10 % FCS (Thermofisher Scientific), 2.5 mg/L amphotericin B (Merck) and penicillin-streptomycin (Merck). Media was changed every 3 days. Controls were glass coverslips (Agar).
For long-term culture, the membrane was fabricated as aforementioned and then mounted onto a grounded water bath as described. Nanoparticles were sprayed onto the membrane using the described parameters followed by cutting and mounting into Scaffdex with the nanoparticle decorated side orientated to be the basal side. Cells were seeded onto the apical side and cultured as mentioned above.
We used histology to determine whether the cells had invaded across the thickness of the membrane. It is important the cells do not invade the entire membrane and culture only on the surface of the membrane. Cell cultured membranes were fixed in 10 % NBF for 15 min and processed for histological staining with Leica TP1020. Following processing, tissues were embedded for sectioning in paraffin wax and stained with Haemotoxylin and Eosin (H&E) using standard protocol, images were taken with Olympus BX60 microscope, 20x objective, no post-processing (
Scanning electron microscopy was used to visualise membrane coverage and cell morphology following cell culture. Cells were fixed in 1.5% Glutaraldehyde (Fluka) for 30 min at 4°C and dehydrated with graded ethanol; 2 × 3 min 50%, 2 × 3 min 70%, 2 × 3 min at 90%, 2 × 5 min at 100%. Hexamethyldisilane (HMDS, Merck) was added for 2 × 5 min and left overnight to evaporate. Samples were mounted on a carbon tab on an aluminium stub, gold sputter coated (Quorum) and imaged using SEM (Quanta FEG250 ESEM) with EHT of 5 kV (
To visualise the presence of RPE characteristic proteins in the ARPE-19 cells cultured on electrospun membranes, we carried out immunofluorescence studies. Cell cultured membranes were harvested after 1 or 3 months of culture, washed with DPBS and fixed in 10% NBF for 15 min. Cells were permeabilised with 1% Triton-X for 10 min, blocked with 10% goat serum for 30 min and incubated with primary antibodies (
Primary antibody stains and the companies from where they were bought with corresponding catalogue numbers.
Anti-Bestrophin/BEST1 antibody | Abcam (ab14927) |
Anti-RPE65 antibody | Abcam (ab13826) |
Anti-LRAT antibody | Abcam (ab166784) |
Anti-PEDF antibody | Abcam (ab180711) |
Anti-CRABP1 antibody | Abcam (ab235838) |
Anti-MERTK antibody | Abcam (ab110108) |
Anti-LAMP2 antibody [H4B4] | Abcam (ab25631) |
Anti-USO1 antibody | Abcam (ab102470) |
Anti-GULP antibody | Abcam (ab236893) |
ZO-1 | Thermofisher Scientific (61-7300) |
Alexa Fluor 488 Phalloidin | Thermofisher Scientific (A12379) |
To assess metabolism of cells cultured on the surface of the membranes, we used a resazurin cell metabolism assay. In this assay, resazurin (an oxidised non-fluorescent blue colored compound) is reduced by metabolically active cells to resorufin (a fluorescent pink product), which is detected via fluorescence measurement. Resazurin solution (5 mg resazurin salt in 40 mL DPBS) (PAA laboratories), was added to each sample (100 μl/ml) and returned to the incubator for 4 h. A 100 μl aliquot of media was taken from each sample and transferred into a black bottomed 96-well plate. Fluorescence was measured at 530–510 nm excitation and 590 nm emission using a fluorescence reader (FLUOstar OPTIMA, BMG LABTECH) (
To assess the ability of the membrane to act as a barrier, we set up a barrier assay using FITC as the detectable dye to traverse the membrane on both a cellular and cell cultured membranes (
Live/dead assays were undertaken to visualise the viability of cells grown on the membranes following culture with and without nanoparticles to assess the effect the presence of degrading nanoparticles had on the cells. Cell cultured membranes were washed in DPBS and following manufacturer’s instructions (molecular probes) incubated with calcein AM and ethidium homodimer-1 for 30 min. The material was placed on glass slides and mounted to coverslips with Vectorshield (Vectorlabs). Slides were viewed using the Zeiss ImagerM1 microscope, x4 objective, Nikon camera, no post-processing (
Three pH measurements were taken from each well using a digital pH meter to assess if broad changes in pH due to nanoparticle degradation over time occurred that could affect cell viability (Mettler-Toledo) (
Release of encapsulated FITC dye from the nanoparticles with time was assessed to determine how long the encapsulated moieties could be sustainably released. Following electrospraying, 10 ml of collected nanoparticles in 0.1% (v/v) isopropanol was kept at 37°C. An aliquot (1 ml) of solution was then collected at each time point and kept at −20°C until the final time point. For collection from cell cultures 1 ml of media was collected at each time point and media changed. A 100 μl aliquot was taken from each sample and transferred into a black bottomed 96-well plate. Fluorescence was measured at 530–510 nm excitation and 590 nm emission using a fluorescence reader (FLUOstar OPTIMA, BMG LABTECH) (
Data are presented as mean (+/− standard error) with
Scaffolds were characterized for adequate mechanical properties to undergo surgical handling and the ultrastructure was characterized to match fibre morphology to mimic the native BM (
Nanofibrous non-degradable PET fibres were produced in a collectable membrane form. SEM micrographs showed the membranes exhibited a randomly orientated fibrous mesh (
SEM micrograph of 17.5 % PET electrospun membrane exhibiting ultrastructure of randomly orientated fibres
The membrane gave an interesting stress vs. strain profile following tensile testing; exhibiting an elastic region (original form is retained following release of load) and a plastic region (membrane deforms indefinitely) before attaining failure exhibiting the UTS (
The membranes gave an interesting stress vs. strain plot, as shown by the representative profile
Treatment of the membrane with cell culture media significantly increased hydrophilicity compared to all other treatment methods, whereas ethanol treatment resulted in an increase in the hydrophilicity compared only to UV treatment. WCA of control membranes averaged at 132.6° which decreased to 85.8° and 118.4° following treatment with media and ethanol, respectively (
Scanning electron microscopy and phalloidin staining showed cells were able to form a monolayer when cultured on the membrane up to 3 months, with microvilli that are phenotypical of RPE cells (
Representative SEM micrographs and corresponding fluorescence images of DAPI/phalloidin stained ARPE-19 cells cultured on PET membranes for 1 month
Representative immunofluorescent images show RPE cells cultured on the electrospun membrane for 1 month stained positively for RPE cell marker proteins and RPE function proteins (green). Cells cultured on the membranes were difficult to image clearly due to the auto-fluorescent nature of the PET fibres. Upon comparison with the positive glass coverslip control, similar staining patterns are apparent, especially noticeable with BEST1, GULP and USO1. DAPI = blue nuclear stain (
H&E stained sections showed the lack of cell invasion in the bulk structure of the membrane, which was the desired characteristic at both 1 week and 3 months’ culture (although some material loss was seen due to the histological processing method in month 3) (
The membrane does not allow invasion of cells through its architecture. H&E staining of cells cultured on the membrane for 1 week
Resazurin assays showed the metabolism of the cells cultured on the membrane was significantly lower compared to positive control in the initial first week of culture, thereafter, the metabolism was not significantly different to the positive control (
Resazurin assay showed metabolism of cells cultured on the membranes was significantly lower in the initial 1 week of culture compared to glass coverslip controls. There was no significant difference in the metabolism of the cells thereafter up to 1 month. Significant difference *
To determine the permeability of the membrane a barrier assay was carried out on acellular membranes and on membranes cultured with ARPE-19 cells for 1 month, as this was when a monolayer had formed. The barrier assay on acellular membranes showed FITC was able to passively diffuse through the membrane in a time dependent manner, taking 80 min for maximum signal to be detected (
Fluorescence barrier assays in the acellular substrate showed passive diffusion that plateaued FITC concentration across the membrane by 70 min
Future experiments to confirm the ability of the RPE cells to perform phagocytosis when seeded on the PET membrane are needed. The presence of the phagocytosis marker MERKT in the ARPE-19 cells cultured on the PET membrane (
Spherical nanoparticles were successfully fabricated with average nanoparticle diameters at 480 nm (+/−252) for PLGA and 59 nm (+/−49) for PGA (
SEM micrographs show PLGA nanoparticles
The nanoparticles did not affect ARPE-19 cell viability or morphology when compared with the membrane only control (
Representative images of live/dead staining (green/red) and corresponding SEM micrographs of ARPE-19 cells after 1 day’s culture
Resazurin assay of ARPE-19 cells plated on membranes with PLGA or PGA nanoparticles decorated on the basal side showed significantly higher metabolism compared to membrane only controls for the first 14 days of culture. There was no significant difference in metabolism of the cells thereafter up to 1 month or compared with coverslip control. Significant difference *
Effect of degrading nanoparticles on pH of ARPE-19 culture media exhibited little change, with PLGA exhibiting a significant decrease compared to membrane only control in week 1, which resolved thereafter. Significant difference *
Although a difference in metabolism of cells and pH of their surrounding environment was observed when cultured with nanoparticles; morphology and protein expression results suggest that the cells were not adversely affected by the membrane/nanoparticle composite.
Short term degradation studies show that over 72 h, both PLGA and PGA nanoparticles degraded releasing FITC into solution (0.1 % isopropanol in dH2O). PLGA nanoparticles exhibited a continuous release profile up to 72 h (
Release of FITC from PLGA and PGA nanoparticles into a solution of 0.1% isopropanol over a short period of time
Long term (up to 28 days) degradation studies showed similar results as the short-term studies with PLGA nanoparticles continuing to sustainably release FITC into solution up to 14 days, decreasing thereafter (
Degradation studies on cell cultured PET membranes decorated with FITC encapsulated nanoparticles on the basal surface, showed PLGA nanoparticles exhibited release of FITC over 28 days. The majority of the dye release was observed up to 7 days for PLGA nanoparticles with fluorescence decreasing thereafter; however, PGA nanoparticles exhibited no significant change until after 28 days, with very little detected fluorescence in comparison to PLGA (
In this article we have described the development of a proof-of-concept composite bioactive membrane that has adequate mechanical properties, permeability and cytocompatibility. The membrane was able to maintain a monolayer of ARPE-19 cells that exhibited phenotypic microvilli structures on its apical surface while providing a sustained release of encapsulated moieties from the membrane’s basal surface.
This composite bioactive membrane exhibits the potential to act as an artificial BM and a potential treatment for atrophic AMD as a novel bioactive cell transplant substrate; to the authors’ knowledge the first of its kind to be developed as a potential treatment for atrophic AMD. There are still many aspects to optimize and analyse, hence, we are now focusing on the optimization and formation of co-axially electrosprayed nanoparticles to encapsulate biologically active moieties to target drusen, and have already achieved the coaxial encapsulation of FITC with Nile red as the outer shell via electrospraying (
The raw data supporting the conclusion of this article will be made available by the authors, without undue reservation, to any qualified researcher.
RM performed the acquisition and undertook the analyses of the data and contributed to the writing of the manuscript. IP and SK provided critical evaluation on the progress of the work and guided the work through clinical relevance. SK also provided feedback on the manuscript. AH prepared, wrote, and revised the manuscript, obtained the funding and is also leading the research.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The authors would like to acknowledge Alison Beckett (University of Liverpool) for her help with the SEM to image the fibrous membrane and Dr. Keith Arnold (Materials Innovations Factory, Liverpool) for his help with SEM in imaging the nanoparticles.
The Supplementary Material for this article can be found online at:
aged related macular degeneration
bestrophin 1
bruch’s membrane
4’,6-diamidino-2-phenylindole
dulbecco’s phosphate buffered saline
fluorescein-5-isothiocyanate
fourier-transform infrared spectroscopy
neutral buffered formalin
poly(ethylene terephthalate)
poly(glycolic acid)
poly(lactic acid-co-glycolic acid)
retinal pigment epithelium
scanning electron microscope
ultimate tensile strength
ultraviolet
vascular endothelial growth factor
water contact angle
young’s modulus.