June 2011
Volume 52, Issue 7
Free
Cornea  |   June 2011
Degradation Studies and Biological Behavior on an Artificial Cornea Material
Author Affiliations & Notes
  • Lívia Santos
    From the CEMUC, Departamento de Engenharia Metalúrgica e dos Materiais, Faculdade de Engenharia, Universidade do Porto, Portugal;
  • Maria Pia Ferraz
    the Faculdade de Ciências da Saúde, Universidade Fernando Pessoa do Porto, Portugal;
  • Yuki Shirosaki
    the Biomaterials Laboratory, Faculty of Engineering, Okayama University, Japan; and
  • Maria Ascensão Lopes
    From the CEMUC, Departamento de Engenharia Metalúrgica e dos Materiais, Faculdade de Engenharia, Universidade do Porto, Portugal;
  • Maria Helena Fernandes
    FMDUP, Laboratório de Farmacologia e Biocompatibilidade Celular, Faculdade de Medicina Dentária, Universidade do Porto, Portugal.
  • Akiyoshi Osaka
    the Biomaterials Laboratory, Faculty of Engineering, Okayama University, Japan; and
  • José Domingos Santos
    From the CEMUC, Departamento de Engenharia Metalúrgica e dos Materiais, Faculdade de Engenharia, Universidade do Porto, Portugal;
  • Corresponding author: Lívia Santos, Rua Dr. Robertos Frias s/n Porto 4200-465, Portugal; livia@fe.up.pt
Investigative Ophthalmology & Visual Science June 2011, Vol.52, 4274-4281. doi:10.1167/iovs.10-6861
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Lívia Santos, Maria Pia Ferraz, Yuki Shirosaki, Maria Ascensão Lopes, Maria Helena Fernandes, Akiyoshi Osaka, José Domingos Santos; Degradation Studies and Biological Behavior on an Artificial Cornea Material. Invest. Ophthalmol. Vis. Sci. 2011;52(7):4274-4281. doi: 10.1167/iovs.10-6861.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: Patients with dry eye syndrome, Stevens–Johnson syndrome, or recurrent transplant rejections are unsuitable to receive a keratoprosthesis. The present work aims at developing a highly biocompatible keratoprosthesis that could be successfully implanted in such patients.

Methods.: Glass-reinforced hydroxyapatite (GRHA) was used to construct this new artificial cornea. To grant the device an adequate porosity, a porogen agent was added in the following percentages: 10, 30, and 50%. Samples were physicochemically analyzed in terms of density, porosity, roughness, degradation, and surface imaging. Biological relevance was assessed by cell culture, MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrasodium bromide) assays, and cell imaging.

Results.: Samples B (30% porogen) and C (50% porogen) were found to be the most porous and also had the roughest topography. Degradation studies showed that under simulated physiologic conditions, no mass loss was found. Conversely, under acidic conditions, a significant mass loss was found. The biological performance of these samples was satisfactory when cultured with human fibroblasts. The MTT assay revealed that samples B and C had greater propensity to cell invasion and proliferation than that of the other tested materials. Cell imaging demonstrated that fibroblasts organized around the pore edges before colonizing it.

Conclusions.: A material with physicochemical and biological characteristics close to an ideal artificial cornea has been fabricated. The GRHA cornea containing 30% porogen is the most promising substitute material due to the biological performance, adequate porosity, and low degradation propensity.

It is estimated that more than 10 million people worldwide develop blindness of corneal origin. 1 The transplant of a natural cornea from a matching donor (penetrating keratoplasty) is the most frequent approach to restore vision to corneal blind patients 2 ; however, the shortness of a biocompatible match and recurrent transplant failure have encouraged the development of artificial corneas, also termed keratoprosthesis. 3 Two important artificial corneas are commercially available (AlphaCor [Argus Biomedical Ltd, Perth, Australia] and Dohlman-Doane Boston KPro [Massachusetts Eye & Ear Infirmary, Boston, MA]). The first device is a single piece, made of polyHEMA, whereas the second one combines a donor cornea with a front and back PMMA plate (recently, the PMMA back plate was replaced by titanium). 3,4 Despite the biocompatibility and ergonomic design of these devices, not all blind patients are potential candidates and the number of extruded implants after short- and mid-term implantation is quite discouraging. 5 15 In fact, melting and extrusion of the implant are the chief issues associated with keratoprosthesis. As such, guidelines were established for the design of artificial corneas: support cell invasion and growth, while exhibiting low levels of degradation. 16,17 Until now, no keratoprosthesis has reached this “gold standard,” partially explaining the rate of extrusion of such ophthalmic devices. It has been demonstrated that bone and tooth exhibits biocompatibility within the ocular environment 18 22 ; thus, it has been speculated that the calcium phosphate materials presently used as bone substitute in dental and orthopedic surgery could be an excellent material to build as a keratoprosthesis. 16,23 This speculation is based on the implantation success rate of the osteo-odontokeratoprosthesis (OOKP). 18 22 This implant is made of biological calcium hydroxyapatite obtained through the chirurgical removal of autologous canine tooth root and surrounding bone. 23 After a period of subcutaneous (SC) implantation, this complex is removed from the skin pocket and a PMMA optical tube is inserted in the center. 23 This structure is sutured to the ocular surface using the tissue previously acquired during SC implantation, and further stabilized with buccal mucosa. 23  
Glass-reinforced hydroxyapatite (GRHA) is a calcium phosphate–based material that has been used with great success in orthopedics and dentistry as bone replacement. 24,25 This material is highly bioactive and promotes a significant cellular growth, culminating in tissue regeneration. 24 31 In the present work, a GRHA material is proposed as a potential candidate for an artificial cornea fringe. Porosity can be controlled by adding different percentages of poly(vinyl alcohol) (PVA) to the GRHA. The fabricated materials were subjected to physicochemical characterizations such as porosity analysis and degradation, as well as a biological evaluation performed in human dermal fibroblast cell cultures. The goal was to develop a slowly degradable, porous, biocompatible material that would comply with the ideal standards established for an artificial cornea. 
Materials and Methods
Preparation of GRHA
GRHA was obtained through the addition of a P2O5-based glass with the chemical composition of 65P2O5–15CaO–10CaF2–10Na2O (mol%) with hydroxyapatite (HA) prepared by chemical precipitation. Glass was added in a proportion of 2.5 wt%. Further details regarding the preparation of GRHA may be found in the literature. 26 28 HA and bioglass were milled to <75 μm before blending for 1 hour in the shaker-mixer (Turbula; Willy A. Bachofen AG Maschinenfabrik, Basel, Switzerland) at 50 rpm. One gram of this blend was weighed and poured into a mold, and disks were prepared by uniaxial pressing at 100 bars. Afterward the disks were sintered at a heating rate of 4°C/min at 1300°C, 1 hour stage at 1300°C, followed by natural cooling inside the furnace. Disks were obtained with diameter 1.20 cm and height 0.27 cm. Disks of HA, prepared by a similar procedure, were used as control in the cell culture experiments. 
Preparation of Porous GRHA
Disks with distinct porosities were prepared by mixing powders of GRHA with PVA, a porogen agent. Materials with different porosities were prepared, where samples A, B, and C had the following compositions: A, 10 PVA-90 GRHA wt%; B, 30 PVA-70 GRHA wt%; and C: 50 PVA-50 GRHA wt%. Both PVA and GRHA powders were mixed through a dry process for 1 hour in the shaker-mixer at 50 rpm. One gram of this blend was weighed and poured into a mold, and disks were prepared by uniaxial pressing at 100 bars. The sintering cycle was comprised of a temperature increase of 1°C/min, up to 500°C, for 4 hours; and 4°C/min, up to 1300°C, for 1 hour. The 4-hour time period was necessary to eliminate all PVA present in the sample, thus leaving voids and subsequently pores. Disc dimensions ranged from 1.2 to 1.3 cm in diameter and 0.60 to 0.64 cm in height. 
X-ray Diffraction
Crystallography was investigated by x-ray diffraction (X'Pert Pro; PANalytical B.V., Almelo, The Netherlands) using a Cu Kα radiation source. Variables were as follows: starting angle, 5.00°; final angle, 70.00°; step size, 0.010°; and scan speed, 0.01°/s. 
Porosity Measurements
Porosity measurements were determined through the Archimedes' principle according to the following equation 32 :   where ρap represents the apparent density of GRHA (3.1) and ρGRHA, the density of either dense GRHA or porous GRHA. Measurements were performed in triplicate for each type of sample. 
The determination of porosity does not disclose any information regarding the pore diameter; therefore, this parameter was estimated by scanning electron microscopy (SEM; FEI Quanta 4000 FEG ESEM; Institute for Critical Technology and Applied Science [ICTAS], Virginia Tech, Blacksburg, VA). Before SEM observations, samples were coated by gold/palladium sputtering in conjugation with argon (SPI Module-Sputter Coater; SPI Supplies Division of Structure Probe, Inc., West Chester, PA) for 120 seconds, giving a film depth of approximately 300 Å. A minimum of five measurements per sample was performed. 
Roughness
Roughness was assessed using a roughness meter (T8000 tester; Hommelwerke GmbH, Schwenningen, Germany), which had a measurement length of 4.80 mm and a cutoff of 0.8 mm. Average roughness (Ra ), maximum roughness depth (R max), and kurtosis (Rku ) were determined. 
Measurements were performed on both sides of each sample in triplicate and two samples of each kind were measured, giving a total of 12 determinations for each type of sample. 
Degradation Studies
Samples were challenged against two types of degradation solutions: one at physiologic pH, to mimic the body fluid under normal physiologic conditions, and the other under acidic conditions (pH 3.0), to mimic the body fluid pH during an inflammatory event. Degradation tests were performed for 1 week; however, physiologic conditions were maintained for an additional period of 4 weeks to achieve more realistic information about long-term outcomes. 
Degradation solutions were conducted according to Standard ISO 10993-14 (“Biological evaluation of medical devices—Part 14: identification and quantification of degradation products from ceramics”). Two degradation solutions were prepared: one with pH 7.4 and the other with pH 3.0. The physiologic solution was prepared by dissolving 13.25 g of Tris(hydromethyl)aminomethane in 500 mL ultrapure water, adjusting the pH to 7.4 using chloridric acid at 1 M at 37°C, then adding ultrapure water for a final volume of 1 L. The acidic solution was prepared by dissolving 21 g citric acid monohydrated in 500 mL ultrapure water, while adding 200 mL of sodium hydroxide (1 M) to ultrapure water until making 1 L of solution; from this solution, 40.4 mL was then mixed with 59.6 mL of chloridric acid (0.1 M). Each sample was soaked in 40 mL of solution and incubated at 37°C with constant agitation. After the incubation period, each disc was filtered, dried to a constant mass, and weighed. Tests were made in triplicate. 
Sample Characterization
Samples were characterized before and after degradation to detect changes in mass and surface profiles using weight loss testing and SEM analysis, respectively. Before SEM observations, samples were coated by gold/palladium sputtering in conjugation with argon (SPI Module-Sputter Coater; SPI Supplies/Structure Probe) for 120 seconds, giving in a film depth of approximately 300 Å. Images were observed under magnifications of ×100 and ×1000, and at least three images were acquired per sample. 
Samples from the long-term degradation assay (pH 7.4) were also subjected to weight loss analysis and SEM observations. In addition, the concentration of calcium ions released into the degradation medium was estimated through high-resolution atomic absorption spectroscopy (ContrAA 700; Analytik, Jena, Germany) equipped with a xenon short-arc lamp (XBO 301, 300W; Gesellschaft für lichttechnische Erzeugnisse mbH, Berlin, Germany) operating in a hot-spot mode as a continuum radiation source was used. An air-acetylene flame was used for atomization of calcium, and all measurements were carried out in triplicate. Analytical solutions containing calcium were prepared in 0.5% (v/v) hydrochloric acid and 1% (m/v) cesium plus 1% (m/v) lanthanum. Plastic bottles and glassware materials were cleaned by soaking in 10% v/v HNO3 for at least 24 hours and rinsed abundantly in deionized water. 
Cell Culture and Seeding onto GRHA
The biocompatibility of either dense as porous GRHA was assessed by exposing the material to a cell culture of human dermal fibroblasts (HFF-1, P7, ATCC). Cryovials containing fibroblasts (7th generation) were heated for 1 minute at 37°C. Under aseptic conditions the content was dipped into a T-Flask containing warm culture medium. The medium composition followed the ATCC recommendations and was prepared with 15% of heat inactivated fetal calf serum (Sigma-Aldrich, Germany), 84% high glucose Dulbecco's Eagle's medium (Sigma-Aldrich, Germany) 0.5% streptomycin and 0.5% penicillin. Fibroblasts were incubated until reaching 80% of confluence, and were then subcultured at a concentration of 3 × 104 cell/mL based on the viable cell exclusion. Viable cells were enumerated by optical microscopy using a Neubauer camera. Dense and porous disks of GRHA were placed into 24-well tissue culture plates and incubated overnight in culture medium. For seeding of samples, 1 mL of the cell suspension was pipetted in each well. Samples were incubated for 14 days at 37°C with 5% CO2 and 80%humidity, and the medium was replaced every 3 days. Cultures performed in dense HA disks were used as control. 
Cell Proliferation/Viability
A reduction of 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrasodium bromide (MTT) to a purple formazan reaction product was used to estimate cell viability/proliferation. MTT was purchased from Invitrogen Life Sciences, Carlsbad, CA. 
For staining, cell cultures were incubated with 0.5 mg/mL of MTT for 3 hours at 37°C with 5% CO2. DMSO was added to dissolve the formazan product, and 100 μL was transferred to a 96-well plate for an absorbance reading at 600 nm. Cell viability/proliferation was evaluated at days 1, 3, 7 and 14. Tests were made in quadruplicate. 
Cell Imaging
Fibroblasts seeded over GRHA disks were imaged by SEM (FEI Quanta 4000, FEG ESEM; ICTAS), after 1 and 14 days of culture. Colonized samples were fixed in 1.5% glutaraldehyde in 0.14 M sodium cacodylate buffer (pH 7.3) for 10 minutes before being dehydrated in graded alcohols, critical-point dried, sputter-coated with gold and finally analyzed. 
Cell cytoskeleton filamentous actin (F-actin) was visualized by treating day 7 and day 14 cells with AlexaFluor 488 phalloidin (1:20 dilution in PBS, 1 hour), and counterstaining with propidium iodide for cell nuclei labeling. Cultures were fixed in 3.7% paraformaldehyde (15 minutes) and cell distribution and infiltration was observed by laser scanning confocal microscopy (LSCM; inverted microscope, Leica DMIRE2; Leica Microsystems GmbH, Wetzlar, Germany). Labeled cultures were mounted in Vectashield and examined. These imaging techniques allowed the visualization of the cell distribution over the materials surface and, more importantly, to verify whether the fibroblasts invaded the porous GRHA. 
Statistical Analysis
Data analysis was performed using a software package (Social Package for Social Sciences; SPSS Inc., Chicago, IL) that uses the one-way ANOVA test with 95% confidence level with Tukey's post hoc. 
Results
X-ray Analysis
Inspection of the dense GRHA spectrum reveals the presence of three distinct phases: HA, β-tri-calcium phosphate (β-TCP) and α-tri-calcium phosphate (α-TCP), as seen in Figure 1. Spectra obtained from porous samples are quite similar between them but different from that found for dense GRHA. In porous samples the peaks matching the soluble phases, α-TCP and β-TCP, are much less pronounced. 
Figure 1.
 
X-ray diffraction spectra obtained for dense GRHA and porous GRHA.
Figure 1.
 
X-ray diffraction spectra obtained for dense GRHA and porous GRHA.
Porosity Measurements
The pore diameter was estimated by taking into account all porous samples. The global average was 109.95 ± 28.85 μm. 
The values of density and porosity are given in Table 1. Samples B and C exhibited the greater porosity percentage (P < 0.05). These samples were prepared with a higher content of PVA than that of sample A, which explains this result. As expected, dense GRHA was the densest and also the least porous material. 
Table 1.
 
Values of Apparent Density, Porosity, and Roughness Determined for Dense GRHA and Porous GRHA
Table 1.
 
Values of Apparent Density, Porosity, and Roughness Determined for Dense GRHA and Porous GRHA
Sample Density (g/cm3) Porosity (%) R a (μm) R max (μm) R ku (μm)
A (10% PVA) 2.69 ± 0.11* 13.21 ± 3.43 7.0 ± 0.9* 63.2 ± 11.7* 20.1 ± 2.6*
B (30% PVA) 2.40 ± 0.03* 22.42 ± 0.85† 6.9 ± 1.3* 57.4 ± 20.1* 21.9 ± 7.9*
C (50% PVA) 2.23 ± 0.03* 28.14 ± 0.95† 8.6 ± 1.9* 76.1 ± 18.7* 20.8 ± 5.0*
Dense GRHA 2.87 ± 0.60 7.37 ± 2.58 2.2 ± 0.5 23.5 ± 11.2 5.8 ± 0.3
Roughness
The results concerning roughness measurements are displayed in Table 1. Average roughness (R a), maximum roughness (R max), and kurtosis (R ku) were analyzed, and R ku is expressed using the peaked distribution. R a and R ku values of porous samples (A, B, and C) were greater than those in dense GRHA (P < 0.05). As expected, sample C exhibited the highest R max compared with that of the other materials (P = 0.012), due to the high degree of irregularities caused by pores. 
Degradation Studies
Low pH Conditions: 1 Week of Degradation.
Porous samples exposed to an acid solution underwent a significant loss of mass (P < 0.05), suggesting degradation of the material. Weight loss was 18.5%, 48.5%, and 55.6%, respectively, for samples A, B, and C, whereas dense GRHA lost 13.5% of its weight (Fig. 2A). Images obtained through SEM (Fig. 2B) showed that porous sample surfaces have a sponge-like appearance, which confirms material degradation. 
Figure 2.
 
(A) Mass of dense GRHA and porous GRHA before and after 1 week of degradation under extreme conditions. (B) SEM appearance of dense GRHA and porous GRHA after 1 week of degradation under extreme conditions. Bar: 1 mm.
Figure 2.
 
(A) Mass of dense GRHA and porous GRHA before and after 1 week of degradation under extreme conditions. (B) SEM appearance of dense GRHA and porous GRHA after 1 week of degradation under extreme conditions. Bar: 1 mm.
Physiologic Conditions: 1 and 5 Weeks of Degradation.
Samples exposed to a physiologic solution did not undergo any loss of mass, after either 1 week or 5 weeks of incubation (Fig. 3). Still, it was suspected that some material dissolution occurred and was confirmed through high-resolution atomic absorption spectroscopy. Table 2 shows that dense GRHA and porous GRHA prepared with 50% PVA (sample C) exhibited the greatest concentration of calcium ions in the test solution (P < 0.05). Results obtained for sample C should be in accordance with its higher porosity and thus greater surface area exposed to the test solution. With respect to the dense GRHA, this result might be in accordance with a greater quantity of the soluble phase's β- and α-TCP (Fig. 1). Images obtained through SEM showed that the surface appearance of sample C changes compared with the control image (before degradation) (Fig. 4). These results showed that, despite the absence of loss of mass, material dissolution actually occurs and that sample C appears to be more prone to it. 
Figure 3.
 
(A) Mass of dense GRHA and porous GRHA before and after 1 week of degradation under physiologic conditions. (B) Mass of dense GRHA and porous GRHA before and after 5 weeks of degradation under physiologic conditions.
Figure 3.
 
(A) Mass of dense GRHA and porous GRHA before and after 1 week of degradation under physiologic conditions. (B) Mass of dense GRHA and porous GRHA before and after 5 weeks of degradation under physiologic conditions.
Table 2.
 
Concentration of Calcium Ions Released into the Physiological Solution (ppm)
Table 2.
 
Concentration of Calcium Ions Released into the Physiological Solution (ppm)
Sample Calcium Ions (ppm)
A (10% PVA) 4.42 ± 0.44
B (30% PVA) 4.61 ± 0.10
C (50% PVA) 6.27 ± 0.13*
Dense GRHA 8.21 ± 0.22*
Figure 4.
 
Pore diameter of porous GRHA estimated by SEM. (A) before degradation, (B) after 1 week, and (C) 5 weeks of degradation under physiologic conditions. Bar: 1 mm.
Figure 4.
 
Pore diameter of porous GRHA estimated by SEM. (A) before degradation, (B) after 1 week, and (C) 5 weeks of degradation under physiologic conditions. Bar: 1 mm.
Cell Culture Studies
The potential hazardous effect of dense and porous GRHA was studied by MTT assay and cell imaging. Figure 5 displays the growing profile of fibroblasts seeded on dense and porous GRHA. Cell proliferation increased with the culture time in all material samples. MTT reduction values on HA (control culture) and dense and porous GRHA were similar at days 1, 3, and 7. However, at day 14, the most porous materials, samples B and C, exhibited significantly higher values of cell viability/proliferation compared with those of the control, dense GRHA, and sample A (P < 0.05). LCSM and SEM observations (Figs. 6, 7) throughout the culture time showed that the porous samples were easily colonized by fibroblasts. The material surface was progressively covered with a cell layer; additionally, fibroblasts were also organized into the pore edges followed by pore colonization. SEM imaging was used to observe a greater number of cells on samples B and C, in agreement with that found in the MTT assay (Fig. 7). 
Figure 5.
 
Fibroblast cell viability/proliferation evaluated by the MTT assay.
Figure 5.
 
Fibroblast cell viability/proliferation evaluated by the MTT assay.
Figure 6.
 
Images of proliferating fibroblasts obtained through laser confocal scanning microscopy. (A) Dense GRHA after 7 days of incubation. (B) Dense GRHA after 14 days of incubation. (C) Sample C after 7 days of incubation. (D) Sample C after 14 days of incubation. (E) Sample C after 7 days of incubation (image with the surface material subtracted). (F) Sample C after 14 days of incubation (image with the surface material subtracted).
Figure 6.
 
Images of proliferating fibroblasts obtained through laser confocal scanning microscopy. (A) Dense GRHA after 7 days of incubation. (B) Dense GRHA after 14 days of incubation. (C) Sample C after 7 days of incubation. (D) Sample C after 14 days of incubation. (E) Sample C after 7 days of incubation (image with the surface material subtracted). (F) Sample C after 14 days of incubation (image with the surface material subtracted).
Figure 7.
 
Images of proliferating fibroblasts obtained through SEM, after 1 and 14 days of incubation.
Figure 7.
 
Images of proliferating fibroblasts obtained through SEM, after 1 and 14 days of incubation.
Discussion
In the present work, attempts were made to produce an artificial cornea fringe, complying with the key features established for this device: controlled degradation profiles, adequate porosity, and a low level of toxicity. Although several groups 33 37 are trying to develop the “gold standard” artificial cornea, we believe the material proposed here, GRHA, is rather promising. Previous studies have evidenced the excellent biocompatibility of GRHA; thus, this material was chosen as a candidate to build an artificial cornea fringe. Sousa et al. 38 implanted a GRHA graft in 11 patients with maxillofacial bone defects. They found excellent bonding between the bone graft and new bone. In orthopedics, GRHA grafts were implanted in the cortical bone of patients who underwent osteotomies. Extensive mature bone formation around the implanted graft and high levels of contact between the bone graft and new bone were found. 25  
Porosity is of utmost importance in the design of an artificial cornea since pores operate as anchorage and proliferation points. In addition, they also function as gateways for the passage of blood vessels and nutrients. In this study, we were able to control porosity through the addition of different percentages of PVA to GRHA. As expected, samples with greater percentages of PVA, samples B and C, were less dense and had greater porosity (Table 1). The greatest porosity percentage was 28.1%, matching GRHA with 50% PVA. The pore dimension is also crucial. Their size should be wide enough to encourage fibroblast ingrowths onto peripheral pores, act as anchorage points, and provide a seal to the ocular stroma. This will prevent future complications, principally extrusion. According to the literature, a keratoprosthesis should have a pore diameter ranging from 50 to 150 μm, to allow cell invasion and proliferation. 3 The pores obtained in this study were subjected to morphologic evaluation, through SEM, and showed an average dimension of 109.95 ± 28.85 μm, thus meeting the desired standards established for an artificial cornea. 
Porosity and pore size are key features of any artificial cornea; however, degradation is also an equally important characteristic. Degradation should be as low as possible, not to compromise the success of the ocular device. In the present study, no loss of mass was observed under physiologic simulated conditions after either 1 week or 5 weeks of degradation (Fig. 3). We have shown that calcium ions were released into the degradation solution, and the surface appearance changed slightly because of this; this was the most notable for sample C. The absence of weight loss under physiologic conditions confirmed we created a material with the lowest degradability possible using this fabrication method. 
Under extreme conditions, degradation occurs in all porous materials. In this case, porosity appears to affect this process since loss of mass was not significant in dense GRHA, yet the most porous materials (samples B and C) were those that underwent the greatest loss of mass (Fig. 2). The extreme solution test is developed as a worst-case environment such as an inflammatory event, where the pH decreases and can virtually reach pH 3.0. The degradation behavior of some medical devices at this pH may be critical to several fields such as dentistry or orthopedics, and therefore is part of the study mentioned in ISO 10993-14 applied to the general medical devices. However, this extreme solution test is not so important for ophthalmic purposes. In this type of applications the local pH will never be as low for such a long period, not only because the body has innate pH-dampening mechanisms to prevent healthy cells and tissues to be damaged by such acid environment but also because of the administration of an anti-inflammatory therapy. 
The biological behavior of the dense and porous GRHA was studied in human dermal fibroblast cell cultures, through the MTT assays and cell imaging. This is because the surgical procedure anticipated for this keratoprosthesis is similar to that used in OOKP, where the prosthesis, before placement into the eye, is implanted in an SC pocket for dermal fibroblast acquisition. Besides, dermal fibroblasts when exposed to the ocular environment acquire characteristics of corneal fibroblasts and have been used to build tissue-engineered corneas. 39  
The present results show that after 14 days of incubation, materials showing the highest percentage of porosity (samples B and C) had the highest score of viable/proliferating fibroblasts, anchoring points, and/or a greater colonizing surface, as evidenced by Table 1. As demonstrated, porous samples are significantly rougher than dense GRHA. In opposition, hydroxyapatite (control culture), dense GRHA, and porous GRHA prepared with 10% PVA (sample A) showed similar behavior in terms of cell proliferation, which should be related with a small percentage of porosity, and thus few anchoring points and a smaller proliferation surface area. 
Bhatia and Yetter 40 reported a correlation between proliferating cells assayed by the MTT assay and visual observation. In this study, we also found that the MTT assay was always corroborated by cell imaging. Thus, at day 14, samples B and C showed increased MTT reaction values, and evidence of a higher number of adherent cells over these materials was also seen with SEM observation. In fact, cell imaging provides a valuable assessment of the material biocompatibility, not only because it supported the results obtained through MTT assay but also because it allows cell morphology to be visualized. Morphology is an essential feature of a biocompatibility evaluation, given that the normal cell shape has to be maintained for optimal cell proliferation and differentiation. 41 LCSM confirmed that fibroblasts growing over GRHA present a normal F-actin cytoskeleton, 41 a structure that controls the shape and surface movements, and provide the molecular basis for many of the mechanical properties of the cytoplasm. 41,42 SEM analyses show it is possible to observe fibroblasts anchoring to the pore edge early in the culture, followed by pore colonization. On day 14, fibroblasts acquired a tissue-like organization presenting the typical elongated morphology and parallel pattern of cell growth. Thus, we have shown GRHA is biocompatible, and performs best when the sample is porous. 
Overall, porous GRHA prepared with 30% PVA (sample B) appears to be a suitable candidate biomaterial for an artificial cornea fringe. This material presents an adequate porosity, shows little dissolution after 5 weeks of exposure to a degradation solution, and has ideal biological behavior. Because this material resists degradation, it should be able to withstand a hostile environment. Further “in vivo” assays will be considered in the near future. 
Footnotes
 Supported, in part, by Fundaçãopara a Ciência e Tecnologia Postdoctoral Grant BPD/43993/2008 (LS) (to ON.2 [O Novo Norte] and QREN [National Strategic Reference Framework]), and by European Union/Fundo Europeu de Desenvolvimento Regional Grant No. 1372 (Project I&DT BIOMAT&CELL).
Footnotes
 Disclosure: L. Santos, Medmat Innovation (F); M.P. Ferraz, None; Y. Shirosaki, None; M.A. Lopes, None; M.H. Fernandes, None; A. Osaka, None; J.D. Santos, P
References
Thylefors B Négrel AD Pararajasegaram R Dadzie KY . Global data on blindness. Bull World Health Organ. 1995;73:115–121. [PubMed]
Albon J . Corneal transplantation and the artificial cornea. J Mech Med Biol. 2003;3:95–106. [CrossRef]
Liu J . Corneal tissue engineering: a review. Biomed Mater Eng. 2008;265.04:1–17.
Hicks C Crawford G Chirila T . Development and clinical assessment of an artificial cornea. Prog Retin Eye Surg. 2000;9:149–170. [CrossRef]
Hollik EJ Watson SL Dart JK Luthert PJ Allan BD . Legeais BioKpro III keratoprosthesis implantation: long term results in seven patients. Br J Ophthalmol. 2006;90:1146–1151. [CrossRef] [PubMed]
Hicks CR Crawford GJ Dart JK . AlphaCor: clinical outcomes. Cornea. 2006;25:1034–1042. [CrossRef] [PubMed]
Alio JL Mulet ME Haroun H . Five year follow up of biocolonisable microporous fluorocarbon haptic (BIOKOP) keratoprosthesis implantation in patients with high risk of corneal graft failure. Br J Ophthalmol. 2004;88:1585–1589. [CrossRef] [PubMed]
Crawford GJ Hicks CR Lou X . The chirila keratoprosthesis: phase I human clinical trial. Ophthalmology. 2002;109:883–889. [CrossRef] [PubMed]
Hicks C Crawford G Chirila T . Development and clinical assessment of an artificial cornea. Prog Retin Eye Res. 2000;19:149–170. [CrossRef] [PubMed]
Legeais JM Ellies PD Dighiero P Renard G . Phacoemulsification in a patient with an artificial cornea (Comment). Arch Ophthalmol. 1999;117:1669. [PubMed]
Legeais JM Renard G . A second generation of artificial cornea (BIOKOP II). Biomaterials. 1998;19:1517–1522. [CrossRef] [PubMed]
Holak SA Holak HM Bleckmann H . AlphaCor keratoprosthesis: postoperative development of six patients. Graefes Arch Clin Exp Ophthalmol. 2009;247:535–539. [CrossRef] [PubMed]
Khan BF Harissi-Dagher M Pavan-Langston D Aquavella JV Dohlman CH . The Boston keratoprosthesis in herpetic keratitis. Arch Ophthalmol. 2007;125:745–749. [CrossRef] [PubMed]
John T . Artificial cornea: surgical use of Boston keratoprosthesis. Ann Ophthalmol (Skokie). 2008;40:2–7. [PubMed]
Leon CR Barraquer JIJr Barraquer JISr . Coralline hydroxyapatite keratoprosthesis in rabbits. J Refract Surg. 1997;13:74–78. [PubMed]
Liu C Paul B Tandon R . The osteo-odonto-keratoprosthesis (OOKP). Semin Ophthalmol. 2005;20:113–128. [CrossRef] [PubMed]
Tay AB Tan DT Lye KW . Osteo-odonto-keratoprosthesis surgery: a combined ocular-oral procedure for ocular blindness. Int J Oral Maxillofac Surg. 2007;36:807–813. [CrossRef] [PubMed]
Ricci R Pecorella I Ciardi A . Strampelli's osteo-odonto-keratoprosthesis. Clinical and histological long-term feature of three prostheses. Br J Ophthalmol. 1992;76:232–234. [CrossRef] [PubMed]
Hille K Grabner G Liu C . Standards for modified osteoodontokeratoprosthesis (OOKP) surgery according to Strampelli and Falcinelli: the Rome–Vienna protocol. Cornea. 2005;24:895–908. [CrossRef] [PubMed]
Fukuda M Nakao A Hamada S . A case of severe Stevens-Johnson syndrome successfully treated by osteo-odonto-keratoprosthesis surgery. Jpn J Ophthalmol. 2005;49:423–424. [CrossRef] [PubMed]
Liu C Herold J Sciscio A Smith G Hull C . Osteo-odonto-keratoprosthesis surgery (Abstract). Br J Ophthalmol. 1999;83:127. [CrossRef] [PubMed]
Mehta JS Futter CS Sandeman SR . Hydroxyapatite promotes superior keratocyte adhesion and proliferation in comparison with current keratoprosthesis skirt materials. Br J Ophthalmol. 2005;89:1356–1362. [CrossRef] [PubMed]
Falcinelli G Falsini B Talone M . Modified osteo-odonto-keratoprosthesis for treatment of corneal blindness. Arch Ophthalmol. 2005;123:1319–1328. [CrossRef] [PubMed]
Gutierres M Sooraj Hussain N Lopes MA . Histological and scanning electron microscopy analyses of bone/implant interface using the novel Bonelike® synthetic bone graft. J Orthop Res. 2006;24:953–958. [CrossRef] [PubMed]
Gutierres M Lopes MA Sooraj Hussain N . Bone ingrowth in macroporous Bonelike® for orthopaedic applications. Acta Biomater. 2008;4:370–377. [CrossRef] [PubMed]
Santos JD Knowles JC Reis R . Microstructural characterization of glass-reinforced hydroxyapatite composites. Biomaterials. 1994;15:5–10. [CrossRef] [PubMed]
Santos JD Jha LJ Monteiro FJ . Surface modifications of glass-reinforced hydroxyapatite composites. Biomaterials. 1995;16:521–526. [CrossRef] [PubMed]
Lopes MA Knowles JC Santos JD Monteiro FJ Olsen I . Direct and indirect effects of P2O5 glass reinforced-hydroxyapatite composites on the growth and function of osteoblast-like cells. Biomaterials. 2000;21:1165–1172. [CrossRef] [PubMed]
Dias AG Lopes MA Trigo Cabral A Santos JD Fernandes MH . In vitro studies of calcium phosphate glass ceramics with different solubility with the use of human bone marrow cells. J Biomed Mater Res A. 2005;74:347–355. [CrossRef] [PubMed]
Ferraz MP Fernandes MH Trigo Cabral A Santos JD Monteiro FJ . In vitro growth and differentiation of osteoblast-like human bone marrow cells on glass reinforced hydroxyapatite plasma-sprayed coatings. J Mater Sci Mater Med. 1999;10:567–576. [CrossRef] [PubMed]
Ferraz MP Fernandes MH Santos JD Monteiro FJ . HA and double-layer HA-P2O5/CaO glass coatings: influence of chemical composition on human bone marrow cells osteoblastic behavior. J Mater Sci Mater Med. 2001;12:629–638. [CrossRef] [PubMed]
Reimbrecht EG Fredel MC Bazzo E Pereira FM . Manufacturing and microstructural characterization of sintered nickel wicks for capillary pumps. Mater Res. 1999;2:225–229. [CrossRef]
Fenglan X Yubao L Xiaoming Y . Preparation and in vivo investigation of artificial cornea made of nano-hydroxyapatite/poly(vinyl alcohol) hydrogel composite. J Mater Sci Mater Med. 2007;18:635–640. [CrossRef] [PubMed]
Sandeman SR Faragher RG Allen MC . Novel materials to enhance keratoprosthesis integration. Br J Ophthalmol. 2000;84:640–644. [CrossRef] [PubMed]
Linnola RJ Happonen RP Andersson OH . Titanium and bioactive-glass-ceramic coated titanium as material for keratoprosthesis. Exp Eye Res. 1996;63:471–478. [CrossRef] [PubMed]
Myung D Koh W Bakri A . Design and fabrication of an artificial cornea based on a photolithographically patterned hydrogel construct. Biomed Microdevices. 2007;9:911–922. [CrossRef] [PubMed]
Viitala R Franklin V Green D Liu C Lloyd A Tighe B . Towards a synthetic osteo-odonto-keratoprosthesis. Acta Biomater. 2009;5:438–452. [CrossRef] [PubMed]
Sousa RC Lobato JV Maurício AC . A clinical report of bone regeneration in maxillofacial surgery using Bonelike® synthetic bone graft. J Biomater Appl. 2008;22:373–385. [CrossRef] [PubMed]
Zhang YQ Zhang WJ Liu W . Tissue engineering of corneal stromal layer with dermal fibroblasts: phenotypic and functional switch of differentiated cells in cornea. Tissue Eng Part A. 2008;14:295–303. [CrossRef] [PubMed]
Bhatia SK Yetter AB . Correlation of visual in vitro cytotoxicity ratings of biomaterials with quantitative in vitro cell viability measurements. Cell Biol Toxicol. 2008;24:315–319. [CrossRef] [PubMed]
Alberts B Johnson A Lewis J . The cytoskeleton. In: Alberts B Johnson A Lewis J Raff M Roberts K Walter P eds. Molecular Biology of the Cell. 4th ed. New York: Garland Science; 2002:907–982.
Stamenovic D . Effects of cytoskeletal prestress on cell rheological behaviour. Acta Biomater. 2005;1:255–262. [CrossRef] [PubMed]
Figure 1.
 
X-ray diffraction spectra obtained for dense GRHA and porous GRHA.
Figure 1.
 
X-ray diffraction spectra obtained for dense GRHA and porous GRHA.
Figure 2.
 
(A) Mass of dense GRHA and porous GRHA before and after 1 week of degradation under extreme conditions. (B) SEM appearance of dense GRHA and porous GRHA after 1 week of degradation under extreme conditions. Bar: 1 mm.
Figure 2.
 
(A) Mass of dense GRHA and porous GRHA before and after 1 week of degradation under extreme conditions. (B) SEM appearance of dense GRHA and porous GRHA after 1 week of degradation under extreme conditions. Bar: 1 mm.
Figure 3.
 
(A) Mass of dense GRHA and porous GRHA before and after 1 week of degradation under physiologic conditions. (B) Mass of dense GRHA and porous GRHA before and after 5 weeks of degradation under physiologic conditions.
Figure 3.
 
(A) Mass of dense GRHA and porous GRHA before and after 1 week of degradation under physiologic conditions. (B) Mass of dense GRHA and porous GRHA before and after 5 weeks of degradation under physiologic conditions.
Figure 4.
 
Pore diameter of porous GRHA estimated by SEM. (A) before degradation, (B) after 1 week, and (C) 5 weeks of degradation under physiologic conditions. Bar: 1 mm.
Figure 4.
 
Pore diameter of porous GRHA estimated by SEM. (A) before degradation, (B) after 1 week, and (C) 5 weeks of degradation under physiologic conditions. Bar: 1 mm.
Figure 5.
 
Fibroblast cell viability/proliferation evaluated by the MTT assay.
Figure 5.
 
Fibroblast cell viability/proliferation evaluated by the MTT assay.
Figure 6.
 
Images of proliferating fibroblasts obtained through laser confocal scanning microscopy. (A) Dense GRHA after 7 days of incubation. (B) Dense GRHA after 14 days of incubation. (C) Sample C after 7 days of incubation. (D) Sample C after 14 days of incubation. (E) Sample C after 7 days of incubation (image with the surface material subtracted). (F) Sample C after 14 days of incubation (image with the surface material subtracted).
Figure 6.
 
Images of proliferating fibroblasts obtained through laser confocal scanning microscopy. (A) Dense GRHA after 7 days of incubation. (B) Dense GRHA after 14 days of incubation. (C) Sample C after 7 days of incubation. (D) Sample C after 14 days of incubation. (E) Sample C after 7 days of incubation (image with the surface material subtracted). (F) Sample C after 14 days of incubation (image with the surface material subtracted).
Figure 7.
 
Images of proliferating fibroblasts obtained through SEM, after 1 and 14 days of incubation.
Figure 7.
 
Images of proliferating fibroblasts obtained through SEM, after 1 and 14 days of incubation.
Table 1.
 
Values of Apparent Density, Porosity, and Roughness Determined for Dense GRHA and Porous GRHA
Table 1.
 
Values of Apparent Density, Porosity, and Roughness Determined for Dense GRHA and Porous GRHA
Sample Density (g/cm3) Porosity (%) R a (μm) R max (μm) R ku (μm)
A (10% PVA) 2.69 ± 0.11* 13.21 ± 3.43 7.0 ± 0.9* 63.2 ± 11.7* 20.1 ± 2.6*
B (30% PVA) 2.40 ± 0.03* 22.42 ± 0.85† 6.9 ± 1.3* 57.4 ± 20.1* 21.9 ± 7.9*
C (50% PVA) 2.23 ± 0.03* 28.14 ± 0.95† 8.6 ± 1.9* 76.1 ± 18.7* 20.8 ± 5.0*
Dense GRHA 2.87 ± 0.60 7.37 ± 2.58 2.2 ± 0.5 23.5 ± 11.2 5.8 ± 0.3
Table 2.
 
Concentration of Calcium Ions Released into the Physiological Solution (ppm)
Table 2.
 
Concentration of Calcium Ions Released into the Physiological Solution (ppm)
Sample Calcium Ions (ppm)
A (10% PVA) 4.42 ± 0.44
B (30% PVA) 4.61 ± 0.10
C (50% PVA) 6.27 ± 0.13*
Dense GRHA 8.21 ± 0.22*
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×