An in-vitro medical device material biocompatibility study using primary cell cultures of rat osteoblasts

International Journal of Scientific Reports Shaama F et al. Int J Sci Rep. 2020 Dec;6(12):485-491 http://www.sci-rep.com pISSN 2454-2156 | eISSN 2454-2164

DOI: https://dx.doi.org/10.18203/issn.2454-2156.IntJSciRep20205020

Original Research Article

An in-vitro medical device material biocompatibility study using

primary cell cultures of rat osteoblasts

Fiayaz Shaama1*, Ann Wilson2, Valerie Stoute3

1Department of Medical Sciences, 2Department of Chemistry, University of The West Indies, Trinidad and Tobago, West Indies 3V.A.S. Consulting, Port-of-Spain, Trinidad and Tobago Received: 31 October 2020 Revised: 12 November 2020 Accepted: 18 November 2020 *Correspondence: Dr. Fiayaz A Shaama, E-mail: fiayaz.shaama@gmail.com Copyright: © the author(s), publisher and licensee Medip Academy. This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

ABSTRACT Background: We questioned if simple chemical methods could be applied as a possible biocompatibility test for biomaterials. Methods: In a qualitative experiment, osteoblasts cells harvested from newly born (3-5) day old sprague-dawley rats were cultivated in growth medium in a controlled environment in the presence of titanium (Ti) (American Elements®), cobalt-chromium (Co-Cr) (Nobilium®), bio-activated rattan wood, orthopaedic bone cement (CEMEX®), implant fixtures (Astra TDC), and resorbable suture material (Poly-Gly-Lac) (VICRYL®). Sample aliquots were withdrawn periodically over 16 days. One and two-dimensional polyacrylamide gel electrophoresis (PAGE) as well as isoelectric focusing (IEF) produced spots that were subjected to enzyme digestion and molecular weight determination. In a follow-up quantitative experiment, the same samples, except for those containing CEMEX and VICRYL, were prepared. The alkaline phosphatase (ALP) activity was monitored continuously using the colorimetric Stanbio® kits. The ALP activities at days 2, 4, 8, 12, and 16, designated as a longitudinal variable, TIME, were analysed, using SPSS V.22, as a mixed ANOVA model with TIME as a repeated measure and the material type as an independent factor. Results: The IEF and second dimension PAGE produced an additional spot for Ti at pH(I) of 5-6. This was identified as IQUB_RAT IQ, a ubiquitin-like domain, molecular weight 2.6 kDa. This method was able to find finite differences in osteoblast activity after initial exposure to a foreign body. Both the ALP activity changes from one day to the next for all materials and the TIME-material type interaction effects were significant (p=0.000). Conclusions: This technique is suitable for use with human cell lines or clones. Experiments like these reduce the need for animal testing. Keywords: Osteoblast, Bioactive wood, Biomarkers, Medical device testing, Alkaline phosphatase

INTRODUCTION

The three basic tests for biocompatibility are the cytotoxicity, sensitization and irritation tests.1,2 Sensitization and irritation tests are performed on animals after proving the device is not cytotoxic to cell lines for example MG-63, MC3T3-E1, SaOs-2 and mouse L929

fibroblasts.3-6 Up to a few years ago there were approximately 3,600 cell lines from over 150 different animal species.7 This poses a serious and daunting challenge when looking for human relevance. The protocols for testing medical devices are given in the ISO standards ISO13485.8 Both animal and in vitro testing are detailed in the ISO10993 series.9

International Journal of Scientific Reports | December 2020 | Vol 6 | Issue 12 Page 485

Shaama F et al. Int J Sci Rep. 2020 Dec;6(12):485-491

New products concurrently come with an increase in animal testing. At the same time there is also a concerted effort to reduce animal testing.10 In terms of using animals for biocompatibility studies, applying the ‘3R’ principal of replacement, reduction and refinement improves the quality of science.11,12 In the early 70’s two high performance methods of separating proteins were available. The first was described by Laemmli.13 With this method, molecular weight was used to separate proteins. The second method by Grownow and Griffith involved iso-electric focusing (IEF), in which the net charge of the protein was used.14 However, the identification of proteins was limited by the detection range of the coomassie dye stain.15 Development of electronic equipment as well as software analysis has improved the method. The current procedure involves the sample preparation, IEF, SDS-PAGE, staining, and molecular weight determination of spots using mass spectrometric analysis. The method used here was adapted from standard protocols of the 2D SDS-PAGE technique.16,17 The Ti, Co-Cr, orthopaedic bone cement and resorbable sutures are materials currently used in various forms in medicine. Bio-activated wood was included in the study because of the recent attention given to wood as a possible raw material for fabrication of artificial bone.18-20

METHODS

Reagent used for cell cultures

One litre of phosphate buffered (PBS) saline was prepared using sterile distilled water (dH2O) containing 8.0 g NaCl, 1.44 g Na2HPO4, 0.24 g KH2PO4, and 0.2 g KCl (10 mM PO43-, 137 mM NaCl, and 2.7 mM KCl). This solution (PBS stock) was adjusted to pH 7.4 using 10 M NaOH and concentrated HCl, then cold filtered into sterile bottles under vacuum using the Nalgene® 20 µM bottle top filter, and stored at room temperature. Hank’s balanced salt solutions (HBSS) were made up with and without calcium and magnesium using sterile dH2O containing 8.0 g NaCl, 0.358 g Na2HPO4, 0.6 g KH2PO4, 0.4 g KCl, 1.0 g glucose, 0.185 g CaCl2 and 0.0978 g of anhydrous MgSO4 (HBSS stock). The pH was adjusted to 7.4 using 10 M NaOH and concentrated HCl, then cold filtered and stored in the refrigerator at 4oC in dark bottles. Both solutions were tested for fungal/bacterial growth by incubating small aliquots in the humidified CO2/Air incubator for 2 weeks. The complete growth medium consisted of a cocktail of the following chemicals: basal cell media 199 or minimum essential media (MEM), purchased from sigma-Aldrich; 10% foetal calf serum (FCS); and 0.2% antibiotic (100 U penicillin/100 µg/ml streptomycin and 0.25 µg/ml amphotericin B), also purchased from Sigma-Aldrich. Collagenase enzyme solution was prepared fresh using HBSS. Collagenase dry powder was purchased from Fisher bioreagents. It was made into solution which contained 0.25% collagenase, 0.5 mM CaCl2 and 0.5 mM MgSO4 freshly made up to 5.0 ml. Trypsin-EDTA

solution, without calcium and magnesium, was purchased from Sigma-Aldrich. This was stored in aliquots of 5 ml between 4oC and 6oC in a refrigerator. The solution consisted of 1.0 mg/ml Trypsin in PBS pH 7.4. Vials were quickly equilibrated to 37oC prior to use. Tyrode’s solution, Ca2+ and Mg2+ free, containing 5 mM EDTA was also purchased from Sigma-Aldrich and stored between 4oC and 6oC in a freezer. This reagent was thawed and equilibrated to 37oC prior to use.

Reagents for electrophoresis

The first dimension IEF consumables were purchased as a kit from BioRad®. The rehydrating buffer (buffer I) consisted of 8 M urea, 4% CHAPS or 3-[(3- cholamidopropyl) dimethylammonio] -1-propanes ulfonate, 0.2% Pharmalyte 3-10, 50 mM Dithiothreitol (DTT) and 0.0002% Bromophenol blue made up in 25 ml double distilled water (ddH2O). The solution of 8 M urea was prepared immediately before use. The Pharmalyte® and 10 mM DTT were added to the rehydration buffer just before loading the sample for the first step of the rehydration process. Bromophenol blue powder was obtained from Sigma. A 1% stock solution was prepared dissolving 100 mg Bromophenol blue in 10 ml ddH2O. The final concentration used in the rehydration buffer was 0.0002%. The equilibrating buffer (buffer II) was used to bind the proteins separated in the first dimension run to SDS (sodium dodecyl sulfate). This buffer consisted of 50 mM Tris-HCl, pH 8.8, 6 M urea, 30% glycerol, 2% SDS and Bromophenol blue. This was prepared by adding together: 10 ml of Tris-HCl, pH 8.8, 72.02 g urea, 69.0 ml glycerol, 4.0 g SDS, 400 µl Bromophenol blue and making up to 200 ml using ddH2O. There were two additional buffers prepared and used prior to the second-dimension run. DTT equilibrating buffer (buffer III) consisted of buffer II with 3% DTT. This was prepared by dissolving 70 mg of DTT in 25 ml of buffer II. It was prepared immediately before use.

The last buffer that was used (buffer IV) consisted of buffer II to which iodoacetamide was added. This buffer was made up using 25 ml of buffer II which contained 2.5% of iodoacetamide. The electrode SDS running buffer used in the second-dimension run was also used to prepare the agarose sealing solution. This buffer was diluted 5X by combining Tris base with other reagents described below. Agarose sealing solution was prepared using 0.5 g agarose, 200 µl Bromophenol blue that was added to 100 ml of the electrode SDS running buffer. This mixture was gently warmed in a 500 ml Erlenmeyer flask until the agarose dissolved. The IPG strips used were 7 cm long by 3 mm wide. The pH range used was 3- 10. The first-dimension gel electrophoresis was performed on a flatbed or horizontal iso-electric focusing system. The total volume loaded with the IPG strips (sample with rehydrating buffer I) was 125 µl. The strips were incubated for 12 hours in the presence of sample solution plus the rehydrating buffer I.

International Journal of Scientific Reports | December 2020 | Vol 6 | Issue 12 Page 486

Shaama F et al. Int J Sci Rep. 2020 Dec;6(12):485-491

Several reagents, which had to be prepared in preliminary steps, were utilized in the formation of the second- dimension gel. First, 100 ml of acrylamide/bis (sigma/aldrich) (30%), prepared using distilled water (dH2O), was filtered through Nalgene® bottle-top filters non-surfactant cellulose acetate membrane 0.2 µm pore size, under vacuum and collected into sterile dark glass bottles. This solution was kept at 4oC in the dark until used. A solution (300 ml) of tris-HCl (sigma/aldrich), concentration of 1.5 M, adjusted to pH 8.8, was prepared using distilled water and also stored at 4oC. Another 100 ml of tris-HCl solution, at a concentration of 0.5 M and adjusted to pH 6.8, was prepared and stored in a similar manner. SDS was prepared at a concentration of 10% using deionized H2O (diH2O) and made up to 100 ml. The sample buffer was made up to 8 ml using the following reagents: 3.8 ml distilled water, 1.0 ml of 0.5 M tris-HCl, pH 6.8, 0.8 ml glycerol, 1.6 ml of 10% SDS, 0.4 ml 2 β-mercaptoethanol and 0.4 ml of 0.05% bromophenol blue. The electrode SDS running buffer was prepared by combining Tris base (15 g/l), glycine (72 g/l) and SDS (5 g/l). This was made up to 300 ml of diH2O and the pH adjusted to 8.3. This was stored at 4oC. Before using, it was warmed to 37oC in the water bath and diluted to 1/5 of the original concentration (300 ml in 1200 ml diH2O).

The concentration of acrylamide used was 12%. It was prepared by combining the following reagents, 33.5 ml dH2O, 25.0 ml of 1.5 M tris-HCL, pH 8.8, 1.0 ml of 10% SDS and 40.0 ml of 30% acrylamide stock. It was de- gassed for approximately 15 minutes before use. In the last step of the procedure, 500 µl of freshly prepared 10% ammonium persulfate (Sigma-Aldrich) and 50 µL of tetramethylethylenediamine (TEMED), from thermo scientific pierce, were added before the resulting liquid was poured into 7.3×8 cm gel electrophoresis plates, separated by a spacer of 0.5 mm.

The stacking gel that formed the comb or wells to hold the samples on top of the separating gel contained a lower concentration of acrylamide. The concentration of acrylamide used in the preparation of this gel was 4%. It was prepared using the following reagents, 6.1 ml dH2O, 2.5 ml of 0.5 M tris-HCl, pH 6.8, 100 µl of 10% SDS and 1.3 ml of 30% acrylamide/bis which was again degassed for 15 minutes. In the last step of the procedure, 50 µl of freshly prepared 10% ammonium persulfate and 10 µl of TEMED were added before the resulting liquid was poured on the top of the separating gel in the glass slabs. A 10-well spacer of 0.5 mm thickness was inserted immediately after the stacking gel was introduced. This effectively formed the wells or sample chambers on the top of the gel slab.

The staining solution used consisted of a 0.1% Coomassie blue R-250 solution made up in 40% methanol and 10% acetic acid, both of which were also purchased from sigma-aldrich. The de-staining solution was made up with 40% methanol and 10% acetic acid.

The potential was set at 120 V and the running time was kept at 1-1/2 hours. Two 24-well cell culture plates were used for the six materials plus one for cells only and a blank. Each of these was set up with replicates of 4.

Protein determination and sample loading

A calibration curve of the sample absorbance at 595 nm vs concentration of protein allowed for the determination of protein in the cultures. A stock solution of bovine serum albumin (BSA) (sigma aldrich) was made up by dissolving 1.0 mg BSA in 1 ml of 0.15 M sodium chloride. The sequential dilution of BSA resulted in solutions ranging from 0.0 µg (pure dH2O) to 400 µg (2.5 X dilution) and was done in duplicate. The Bradford 5x reagent was diluted 1 to 1. The calibration curve and unknown protein quantitation was done at the same time. Following this, an equal amount of protein was loaded onto the gels which run in the first dimension. The amount of protein loaded for the first dimension was 150 µg. A volume of sample containing 150 µg was added to the corresponding volume of rehydration buffer to make up a total volume of 125 µl. This was introduced into the well after which the IPG strips were placed gel-side down to start the rehydration process.

Preparation of test materials

All the test materials used were shaped as discs of comparable dimension and fitted into wells of a 24-cell culture template. The exception was the resorbable suture material (Poly-Gly-Lac) (VICRYL®). The entire length (16 mm 3-0 Coated Vicryl TM, Ethicon®) was rolled into a circle 8.0 mm in diameter and fitted into the well of the experimental template. Titanium discs used were purchased from (American elements, Merelex Corp.). The Co-Cr alloy was purchased as ingots (Nobilium Inc.) and casted into discs. The PMMA used was CEMEX® (Tecres® lot. AA9382). This is a type of orthopedic bone cement. It was obtained from a local hospital. It was sold as a packet containing 20.0 g powder along with and separate from 8.35 ml of a liquid component. The powder and liquid were mixed according to the manufacturer’s instruction and molded into discs before complete polymerization. The bio-activated wood was prepared using the protocol given for phosphorylation of microcrystalline cellulose Patent: US2759924.21 The activated wood discs and other test materials were all preconditioned before use. This was done by incubating them in the growth medium in a humidified incubator (5%/95% CO2/Air at 37oC) and checked for bacterial/fungal growth within 2 weeks.

Animal model

Sprague dawley rats were obtained from Charles Rivers Laboratories® USA. Newly born rats were used to obtain osteoblast cells. The method used here involved the sequential enzymatic digestion of the calvaria as previously described.22 The details of weight and sex of

International Journal of Scientific Reports | December 2020 | Vol 6 | Issue 12 Page 487

Shaama F et al. Int J Sci Rep. 2020 Dec;6(12):485-491

the rats used are given in Table 1. The harvested cells were collected as a pellet and re-suspended in 10.0 ml of growth medium by gentle vortex. The growth medium consisted of MEM containing 10% FCS and 0.2% antibiotics (100 units of penicillin, 100 μg/ml streptomycin, and 0.25 µg/ml amphotericin B). Freshly re-suspended cells were then distributed equally into 24 wells containing titanium, cobalt-chromium, PMMA and wood. A blank with growth medium only was included. A cell count (using Trypan Blue solution) was done to estimate how many cells were introduced into each experimental well. The average number of cells introduced in each well was 6.25×105.

Table 1: The weights and sex of the 6 rats used in the

2-D SDS-Page Experiment.

Cell culture template and sample preparation

Four replicates of 1.0 ml culture medium were used for each test material. These were pipetted into 1.8 ml Nalgene® cryogenic vials then frozen in the refrigerator. After 2 hours they were removed and placed in liquid nitrogen. The ALP assay was performed manually. The vials were removed from the cryogenic chamber and thawed on the bench top for 10 minutes. An aliquot of 500 µl from each replicate was pipetted into 2.0 ml Eppendorf tubes and centrifuged at 1000 g for 10 minutes. After centrifugation 20 µl was pipetted from each replicate and added to clean test tubes containing 1.0 ml buffer reagent from the Stanbio® kit and incubated at 37oC in the water bath for 5 mins. To this 250 µl of substrate was added and the timer started. This liquid was transferred to the optical cuvette and the absorbance at 405 nm was recorded every minute for 4 mins. The liquid was then discarded and the procedure repeated for all samples. The ALP measurements were done on days 2, 4, 8, 12 and 16 days at which time the experiment was discontinued. Samples were also collected in a similar manner on the 9th day for the 2D experiments.

Statistical analysis

Quantitative data from the ALP activity measurements for different materials on each of five days after incubation were analysed with SPSS V.22, using a mixed model analysis of variance (ANOVA) routine with ‘time after incubation’ of the biomaterial (with or without the cell cultures) designated as a repeated measure (TIME) and the type of biomaterial as an independent predictor (Material). ALP activity was the outcome (dependent variable) used for the effectiveness of biocompatibility under the different experimental combinations of the categories of TIME and Material. This SPSS routine for

repeated measures (‘within subjects’) ANOVA offers three corrections, for situations when the assumption of sphericity is not met. All of these corrections involve adjusting the p value upwards so that it becomes more difficult to meet the inference of significance (p<α). These are the Greenhouse-Geiser, the Huynh-Feldt, and the lower bound corrections. The significance level was set at p<0.05. Least significant difference (LSD) post hoc tests were conducted to do pair-wise comparisons of effectiveness among the biomaterials used.

RESULTS

The separation of proteins found in media using one dimensional polyacrylamide gel electrophoresis

The images of the various gels were photographed and uploaded onto the computer. The software program SynGene® was used to identify the presence of bands and relative molecular weights based on the standard wide molecular weight range Sigma® (No: 4036) molecular weight ladders. There is a limited number of inferences which can be drawn for qualitative experimental techniques. That being said, when all materials were matched on one gel at day three, for the implant fixture, wood, Ti and VICRYL® there was no protein band in the 51-59 kDa molecular weight range. The media which contained heat-cured PMMA, cobalt- chromium, and cells only produced similar numbers of protein bands with equally matching molecular weights. At day 7, there was greater uniformity in protein distribution seen for all materials. However, the number of bands appearing in the gel for each material dropped from 5-6 bands at day 3 to 4 bands for each material tested. At day 10 the pattern remained essentially consistent with that at day 7 in four systems -the medium only, PMMA, Ti and VICRYL®. Four bands were seen at approximate molecular weights in the range 70-126. kDa. The samples of culture medium taken at day 10 from wells containing cobalt-chromium, wood, and the implant fixture no longer showed a protein band corresponding to 24-25 kDa. Similar observations were found in the gels ran for samples of medium, extracted on different days from wells with individual test materials. This was true for all the materials. For example, in the gels where samples from the wells containing the implant fixture, wood, Ti and VICRYL® were analysed individually, there was an absence of protein bands in the 51-59 kDa range on day three, as reported above when all the materials were matched on one gel. At days 7 and 10, the protein bands produced were also comparable to those found in gels when all materials were matched on one gel.

Two-dimensional gel electrophoresis, ALP activity and mass spectrum analysis

The gel, which ran samples of medium (day 9) in the cells containing Ti, showed one additional spot at the isoelectric pH, pH(I) 5-6. This was in a different or

International Journal of Scientific Reports | December 2020 | Vol 6 | Issue 12 Page 488

Shaama F et al. Int J Sci Rep. 2020 Dec;6(12):485-491

unmatched position when all the gels were compared together. The spot was excised and subjected to mass spectrometric analysis. The molecular mass was estimated to be 2.6 kDa. The online mass spectrum software MATRIX SCIENCE® was used to upload the data that was generated for this protein. The software program estimates that a protein score of greater than 52 is a match. The software identified this protein as IQUB_RAT, IQ and ubiquitin-like domain-containing protein OS=Rattus.