Nitric oxide (NO) is a mediator involved in bone regeneration. We therefore examined the effect of the novel NO donor, S-nitroso human serum albumin (S-NO-HSA) on bone formation in a rabbit calvaria augmentation model.
Circular grooves (8 mm diameter, two per animal) were created by a trephine drill in the cortical bone of 40 rabbits and titanium caps were placed on the rabbit calvaria bone filled with a collagen sponge soaked with either 100 μL S-NO-HSA (5%, 20%) or human albumin (5%, 20%). After 4 weeks the titanium hemispheres were subjected to histological and histomorphometric analysis. Bone formation and the volume of the residual collagen sponge were evaluated.
S-NO-HSA treatment groups had a significantly higher volume of newly formed bone underneath the titanium hemispheres compared to the albumin control groups (5%: 15.5 ± 4.0% versus 10.6 ± 2.9%; P < 0.05; 20%: 14.0 ± 4.6% versus 6.0 ± 3.8%; P < 0.01). The volume of residual collagen sponge was also significantly lower in the S-NO-HSA groups compared to the control groups (5%: 0.4 ± 0.5% versus 2.6 ± 2.4%; P < 0.05 and 20%: 1.5 ± 2.7% versus 13.0 ± 18.7%; P < 0.01).
This study demonstrates for the first time that S-NO-HSA promotes bone formation by slow NO release. Additionally, S-NO-HSA increases collagen sponge degradation.
Bone regeneration is initiated immediately after injury and follows a conserved sequence of cellular events ultimately leading to formation of new mineralized tissue. Successful bone regeneration is required in spontaneous fracture healing but also in therapeutic interventions e.g. osseointegration of dental implants, graft consolidation, and reconstruction surgery. Progress has been achieved in the development of pharmacological therapies making clinical success more predictable and faster, in particular in patients with a compromised regenerative capacity. The most successful pharmacological therapies have been developed based on the mechanistic understanding how particular signalling molecules orchestrate the conserved sequence of cellular events of bone regeneration.
Nitric oxide (NO) is a labile signalling molecule which is produced from l -arginine and oxygen and has pleiotropic functions. All three NO synthase isoenzymes generating NO seem to modulate fracture healing: neuronal NO synthase (nNOS), inducible NO synthase (iNOS) and endothelial NO synthase (eNOS). Insights into the functional role of NO during bone regeneration have been gained from animal models that fail to express iNOS and by studies with inhibitors of NO synthases. It has been reported that systemic administration of l -arginine can improve fracture healing in rats indicating that the supply of l- arginine may be a limiting step in NO synthesis in pathophysiological situations. Also local application of the NO-donor NONOate linked to chitosane has been shown to improve callus diameter, which is a parameter of fracture healing in mice. NO-donors are thus potential candidates in the development of therapies to improve bone regeneration.
S-nitroso human serum albumin (S-NO-HSA) is a novel high-molecular-weight S-nitrosothiol NO-donor. Compared to low molecular weight S-nitrosothiols, S-NO-HSA has a prolonged initial half-life of approximately 15–20 min. Systemic administration of S-HO-HSA has shown to have protective effects in various ischaemia/reperfusion injury models, and prevents platelet aggregation and leucocyte–endothelial cell adhesion. Our hypothesis is that NO slowly released from S-NO-HSA can improve bone formation. The difference in release kinetics between high and low molecular weight S-nitroso-thiols can be shown by the effects on blood pressure. Fracture healing involves the coordinated action of endochondral and intramembraneous ossification. However, osseointegration of dental implants and graft consolidation almost exclusively involves intramembraneous ossification. At present it is unknown whether local application of S-NO-HSA can boost bone regeneration by intramembraneous ossification.
In this study we intended to test S-NO-HSA in a model where titanium caps are placed on the rabbit calvaria bone. The titanium caps were filled with collagen fleece saturated either with S-NO-HSA or human serum albumin (HSA). Due to the use of HSA as control by an otherwise identical procedure occurring differences in bone formation can only result from NO released by S-NO-HSA. Bone regeneration exclusively originates from the surface of the calvaria. The outer aspect of the cortical bone is penetrated thereby providing a connection to the marrow elements that drive bone regeneration. The space underneath the hemispheres is protected from the periosteum and the surrounding soft tissue, following the concept of guided bone regeneration. Based on histological and histomorphometric analysis we tested the effect of local application of S-NO-HSA on bone regeneration in rabbits.
Materials and methods
Animal characteristics and ethics
All animals were treated according to the guidelines of animal care. The model was described in detail elsewhere. The study was carried out on forty male New Zealand White rabbits (Charles River GmbH, Sulzfeld, Germany) with a body weight of 3000–3500 g in accordance with institutional guidelines. The animals had free access to water and food except 12 h prior to surgery (food withdrawal).
Preparation of S-NO-HSA
The preparation of S-NO-HSA has recently been described in detail. In brief: 20% human serum albumin (HSA; Baxter, Vienna, Austria) was processed to yield a maximal free thiol group (-SH) at position Cys-34 (-SH > 0.8 mol/mol protein). Intermolecular disulfides (mixed disulfide) were disassembled prior to nitrosation. The starting material (20% HSA) was reduced by mercaptoethanol (10–20-fold molar excess; buffer: sodium phosphate (1 mmol/L), ethylenediaminetetraacetic acid (2 mmol/L) and sodium chloride (150 mmol/L) adjusted to pH = 6.0–6.2 with hydrochloric acid (HCl); 12–48 h at 4 °C under nitrogen) and purified by means of gel-permeation chromatography (TSK-HW40F; mobile phase: H 2 O). Thiol nitrosation was induced with sodium nitrite at a ratio of 1:1 to 1:1.5 of freely available thiol groups to nitrite in 0.2 mol/L HCl (pH = 1.5–2.5) for 30 min at 25 °C. After neutralization with 1 mol/L sodium hydroxide, S-NO-HSA was purified by gel-permeation chromatography (TSK-HW40F; mobile phase: H 2 O) and lyophilized. Salt-free HSA was also prepared from 20% HSA by means of gel-permeation chromatography (TSK-HW40F; mobile phase: H 2 O). Immediately before surgery S-NO-HSA and HSA were dissolved in physiological saline to yield 5% and 20% solutions, respectively, and subjected to sterile filtration.
Operations were performed under initial sedation with ketamine hydrochloride (15 mg/kg Ketalar ® , Pfizer AG, Zürich, Switzerland) and xylazine-hydrochloride (1.5 mg/kg Rompun ® , Bayer, Leverkusen, Germany). The animals were intubated and lung ventilation was maintained with N 2 O/O 2 (FI O 2 = 0.35) and isoflurane (1–2 vol%; Forane ® , Abbott, Vienna, Austria) at a tidal volume of 15–20 ml/kg and at a rate of 30–35 cycles/min. After local anaesthesia with 1% lidocaine (Xylanest ® Purum 1%; Gebro Pharma GmbH, Fieberbrunn, Austria) a cutaneous flap was created following a midsagittal incision. Subsequently two circular grooves were created using a trephine drill (8 mm outer diameter; Stoma, Emmingen-Liptingen, Germany) under permanent irrigation with sterile saline on both sides of the sagittal suture. The drill was immersed until the point wherein the jagged edges barely cut through the cortical bone. Nine holes were drilled by a rose bur (1 mm outer diameter; Komet Group GmbH, Besigheim, Germany) also under permanent irrigation with sterile saline to induce bleeding from the marrow space within the circumference of the circle. Machined and polished hemispheres of titanium with an outer diameter of 8 mm and thickness of 0.2 mm (Blue Cat, Vienna, Austria) were filled to the brim with three layers of a collagen fleece (Lyostypt ® , Braun, Tuttlingen, Germany) saturated with 100 μL of either S-NO-HSA or HSA solutions (5% or 20%). The animals were divided into four groups ( n = 10 per group). The hemispheres were fixed by press-fit on the two circular grooves on the calvaria surface. Following incision of the periosteum the flaps were sutured. After four weeks, animals were sacrificed after induction of deep anaesthesia followed by a lethal intravenous overdose of sodium thiopental (Thiopental, Sandoz GmbH, Kundl, Austria) and the biopsies were taken.
The calvarias were removed and immediately placed in a buffered formaldehyde solution. The histological preparation has been previously described in detail. In brief, the specimens were dehydrated in ascending grades of ethanol and embedded in 2-hydroxyethyl methacrylate-based polymers (Technovit 7200, Heraeus Kulzer GmbH, Wertheim, Germany). Undecalcified ground sections with a thickness of about 20 μm were prepared and stained with the Levai Laczko stain. The plane of the ground thin sections was oriented perpendicularly to the ectocranial surface of the parietal bone, cutting the hemispheres exactly in half.
Using a microscope (Nikon Microphot FXA; Nikon, Tokio, Japan), ground sections were digitally photographed (1 mm = 452 pixels and 1 pxl = 2.21 μm) and the histologic structures of interest were automatically detected and histomorphometrically measured with a morphometry software (Definiens Developers 6.02; Definiens, Munich, Germany) thereby determining the ‘Bone volume per tissue volume’ (BV/TV) and the ‘Collagen volume per tissue volume’ (CV/TV) in the tissue filled area inside the cap.
Estimation of NO release from titanium caps in vitro
S-NO-HSA and HSA containing caps ( n = 3 per concentration) were prepared exactly according to the description for the in vivo experiments. Each prepared titanium cap was placed into pre-swollen dialysis coil (Membra Cell MD 10 14× 100 CLR Visking ® , Carl Roth GmbH, Karlsruhe, Germany) and secured with clips to minimize the external space. Each so prepared cap was inserted into a well (six well plate) filled with 6 mL physiological saline. The plates were then incubated at 37 °C for 4 weeks in an incubator (Heracell 240 CO 2 incubator, Thermo Scientific, Waltham, MA, USA) and 400 μL protein-free supernatant from each well were taken each week for nitrite/nitrate and protein determination. The volume loss (400 μL) was substituted with physiological saline and the dilution was considered in the calculations.
No protein in the supernatant was the prerequisite for estimation of NO release. Protein-free ensured that no protein leakage (specially relevant for S-NO-HSA) occurred in the testing system. Protein was determined with the BCA method (BCA Protein Assay, Pierce, Rockford, IL) and nitrite/nitrate concentrations were determined with HPLC and fluorometric detection according to the method of Li et al. Nitrate in the samples was converted to nitrite by a Cu/Cd catalyst. Nitric oxide is oxidized to nitrite and nitrate and the quantification of these stable anions can be used to estimate the amount of released nitric oxide. The continuous accumulation of nitrate/nitrites (S-NO-HSA caps) in the supernatant of the wells was measured each week and expressed in nmol. The experiments per concentration were performed in triplicate.
The t -test was used to test for differences in BV/TV, and the Mann–Whitney test for differences in CV/TV. All data analyses were performed using SPSS version 14.0 (SPSS, Inc., Chicago, IL). Bonferroni post hoc test for multiple testing was applied. P -values < 0.05 were considered statistically significant.
Bone formation was restricted to the lower third of the hemispheres, which represent the regions immediately adjacent to the calvaria bone. Inflammatory cells were observed around the new formed tissues in all groups. Inflammation outside the hemispheres was existent in a few ground sections. Two biopsies were excluded from histomorphometric analysis due to inflammatory resorption. Remaining collagen carriers were clearly visible in albumin control but not in the S-NO-HSA groups ( Fig. 1 A and detailed view: Fig. 1 B). Higher microscopic magnifications revealed that bone formation mainly occurred in the region immediately adjacent to the titanium hemispheres. Newly formed bone showed the characteristic signs of woven bone. Independent of the treatment, the absolute amount of newly formed bone under the capsule was low in all groups.
Standard histomorphometric findings
Serial histomorphometric measurements of bone formation ( Fig. 2 A) revealed a significant increase in the relative bone volume under the hemisphere in the group receiving 20% S-NO-HSA compared to 20% HSA (S-NO-HSA: 14.0 ± 4.6% vs. HSA: 6.0 ± 3.8%, respectively; P < 0.01). The corresponding comparison of 5% of S-NO-HSA and 5% HSA as well reached the level of significance (S-NO-HSA: 15.5 ± 4.0% vs. HSA: 10.6 ± 2.9%, respectively; P < 0.05).