Peptide:glycosaminoglycan hybrid hydrogels as an injectable intervention for spinal disc degeneration

Peptide and glycosaminoglycan hybrid gels undergo self-assembly and result in tuneable mechanical properties with suitability for intradiscal treatments.

Transmission electron microscopy, TEM P 11 -9 and P 11 -9:GAG samples were prepared in DPBS (Dubeccos phosphate buffered saline, Sigma Aldrich), solution containing 0.04% NaN 3 , (Sodium azide, Sigma Aldrich, 99%)to prevent bacterial growth. The samples were vortexed for 20 seconds, and then sonicated for 30 minutes. Solution pH was measured and adjusted to 7.4 ± 0.3. Any adjustment of pH, if necessary, was made with μL volumes of 0.1 to 1 mol dm -3 HCL (Hydrochloric acid, Fisher Scientific, analytical reagent grade, 37% wt diluted with purified water), and/or 0.1 to 1 mol dm -3 NaOH (Sodium hydroxide pellets, Fisher Scientfic, analytical grade, dissolved in and diluted with purified water). Solution vials were closed, sealed with PTFE tape, and then warmed to approximately 80 o C to maximise peptide solubility. Following warming GAG powder was added to the peptide solutions and the mixture was vortexed to ensure complete mixing The resulting gels were left to equilibrate for approximately four and a half months.
For P 11 -4, P 11 -8 and P 11 -12, samples with and without GAG prepared for FTIR were used, these were left to equillibriate for 4 days prior to TEM grid preparation.
The gels were then diluted (20 µl of sample in 60 µl D2O) minutes prior to addition to the TEM grids. Glow discharged carbon film coated copper TEM grids (Agar Scientific) were touched onto the peptide and peptide:gag solutions for one minute. The grid was air-dried and then negatively stained with uranyl acetate solution (2% w/v) for 20 seconds and air-dried.
TEM images for P 11 -4, P 11 -8 and P 11 -12, samples with and without GAG were obtained using a Jeol JEM1400 transmission electron microscope operating at 120 kV accererating voltage. TEM images of P 11 9 with and without GAG were obtained using a Philips CM10 electron microscope operating at 80 kV accelerating voltage.

Rheometry
Peptide and peptide:GAG samples were prepared in DPBS (Dubeccos phosphate buffered saline, Sigma Aldrich), solution containing 0.04% NaN 3 (sodium azide, Sigma Aldrich, 99%). The samples were vortexed for 20 seconds, and then sonicated for 30 minutes. Solution pH was measured and adjusted to 7.4 ± 0.3. Any adjustment of pH, if necessary, was made with μL volumes of 0.1 to 1 mol dm -3 HCL (hydrochloric acid, Fisher Scientific, analytical reagent grade, 37% wt diluted with purified water), and/or 0.1 to 1 mol dm -3 NaOH (Sodium hydroxide pellets, Fisher Scientfic, analytical grade, dissolved in and diluted with purified water). Solution vials were closed, sealed with PTFE tape, and then warmed to approximately 80 o C to maximise peptide solubility. Following warming GAG powder was added to the peptide solutions and the mixture was vortexed to ensure complete mixing The resulting gels were left to equilibrate at room temperature (P 11 -4 20 days, P 11 -8 1.5 months, P 11 -9 one month and P 11 -12 two months).
Rheometry measurements used a Malvern Kinexus Pro rheometer with a cone-plate geometry (cone angle: 1 O , diameter: 50 mm, gap: 0.033 mm). All tests were performed at 25 o C, utilizing a solvent trap and the atmosphere within was kept saturated to minimize evaporation of the peptide samples. The geometery was lowered into position and samples incubated for 15 min.
To ensure measurements were made in the linear viscoselastic regime, amplitude sweeps were performed in a shear strain controlled mode from 0.01-100% at 1 Hz and 20 Hz.
The dynamic moduli of the hydrogels were measured as a frequency function with the sweeps carried out between 1 and 20 Hz.

Bovine disc model preparation
Bovine tails were harvested from calves aged less than 30 months at a local abattoir. To avoid potential damage to the tissue, the discs were stored at 2-8 o C prior to experimentation. The tails were cleaned and the spinous processes removed. The discs (C1-C6) were then excised via a transverse cut between the disc and the most cranial cartilage end plate and via a transverse cut through the vertebra, leaving ~10 mm of bone attached to the most caudal side of the disc.
They were placed in monosodium citrate at pH 7.4 for 20 minutes to remove excess blood and through swelling pressures help differentiate the nucleus pulposus from the annulus fibrosus tissue. The discs were then removed from the solution and the nucleus pulposus tissue was excised. The disc and nucleus tissue were weighed prior to and post removal. The discs were then attached to lightly sanded artificial Perspex endplates (2x40x40 mm) using the adhesive Loctite 3090 (Henkel).
For each GAG or Peptide:GAG ratio investigated, discs were chosen from more than one tail over the range of C1 to C6 to try to eliminate variables in disc size, natural GAG levels and tail health. For each group studied, a control of no GAG injection to a denucleated disc was used to determine the average natural GAG leakage over all six tails.

Assessment of GAG leakage through the annulus fibrosus and injection hole
GAG only and peptide solutions were prepared in PBS (phosphate buffered saline solution) and vortexed for 1 min. A total of 250 μl of solution was injected into a disc using a 25 gauge needle (outer diameter 0.51 mm) and syringe with a second 25 G needle as an air hole. For the peptide:GAG augmented discs following an initial injection of 125 μl of the peptide solution, 125 μl of the GAG was injected through the second 25 G needle, retaining the needle used for the peptide injection as a new air hole.
Following injections, the discs were placed in 30 ml of PBS on a orbital shaker for 48 hours. 3 ml aliquots were removed from the surrounding PBS at 24 hours and 48 hours. The 3ml PBS aliquots were analysed via a standard DMMB (dimethylmethylene blue) assay using chondroitin-6sulphate sodium salt (Sigma Aldrich) to construct the standard curve.
Values are reported as means ± standard error of the mean (SEM), n=3. Results of DMMB assays were analyzed by one-way ANOVA, followed by Tukey's multiple comparison test and were assessed for their statistical significance via paired t-tests (P values < 0.05 were considered as statistically significant).

Static compressive loading
Bovine caudal discs were obtained and prepared as per the above GAG leakage study.
Following preparation, all discs were sealed in individual plastic bags with PBS soaked tissue paper to prevent samples drying and stored at 2-8 o C overnight. The most caudal vertebra section of each sample was then cast in 70 mm diameter PMMA cement (polymethylmethacrylate, WHW Plastics) to produce a flat surface that was parallel to the Perspex endplate. Once set, the samples were stored in individual sealed plastic bags containing PBS soaked tissue paper at 2-8 o C until testing.
All specimens underwent static axial compressive loading from 0 to 9 kN using displacement control at a low load rate of 1 mm/min. The experiments were carried out on an Instron 3366 materials testing machine with a 10 kN load cell (Instron, UK).
It should be noted that at the start of each test, the fixture on the crosshead of the materials testing machine was brought into contact with the Perspex endplate until a load of 0.3 N was recorded. At this point, the displacement and load were zeroed and the test was started.
From the resulting load/displacement plots the normalised stiffness was calculated by multiplying by the specimen height and dividing by its cross-sectional area.
Values are reported as means ± standard error of the mean (SEM), n=6. Results were analyzed by one-way ANOVA, followed by Tukey's multiple comparison test and were assessed for their statistical significance via paired t-tests (P values < 0.05 were considered as statistically significant).