Osteogenic differentiation and reconstruction of mandible defects using a novel resorbable membrane: An in vitro and in vivo experimental study

To evaluate the cellular response of both an intact fish skin membrane and a porcine-derived collagen membrane and investigate the bone healing response of these membranes using a translational, preclinical, guided-bone regeneration (GBR) canine model. Two different naturally sourced membranes were evaluated in this study: (i) an intact fish skin membrane (Kerecis Oral (R), Kerecis) and (ii) a porcine derived collagen (Mucograft (R), Geistlich) membrane, positive control. For the in vitro experiments, human osteoprogenitor (hOP) cells were used to assess the cellular viability and proliferation at 24, 48, 72, and 168 h. ALPL, COL1A1, BMP2, and RUNX2 expression levels were analyzed by real-time PCR at 7 and 14 days. The preclinical component was designed to mimic a GBR model in canines (n = 12). The first step was the extraction of premolars (P1-P4) and the 1st molars bilaterally, thereby creating four three-wall box type defects per mandible (two per side). Each defect site was filled with bone grafting material, which was then covered with one of the two membranes (Kerecis Oral (R) or Mucograft (R)). The groups were nested within the mandibles of each subject and membranes randomly allocated among the defects to minimize potential site bias. Samples were harvested at 30-, 60-, and 90-days and subjected to computerized microtomography (mu CT) for three-dimensional reconstruction to quantify bone formation and graft degradation, in addition to histological processing to qualitatively analyze bone regeneration. Neither the intact fish skin membrane nor porcine-based collagen membrane presented cytotoxic effects. An increase in cell proliferation rate was observed for both membranes, with the Kerecis Oral (R) outperforming the Mucograft (R) at the 48- and 168-hour time points. Kerecis Oral (R) yielded higher ALPL expression relative to Mucograft (R) at both 7- and 14-day points. Additionally, higher COL1A1 expression was observed for the Kerecis Oral (R) membrane after 7 days but no differences were detected at 14 days. The membranes yielded similar BMP2 and RUNX2 expression at 7 and 14 days. Volumetric reconstructions and histologic micrographs indicated gradual bone ingrowth along with the presence of particulate bone grafts bridging the defect walls for both Kerecis Oral (R) and Mucograft (R) membranes, which allowed for the reestablishment of the mandible shape after 90 days. New bone formation significantly increased from 30 to 60 days, and from 60 to 90 days in vivo, without significant differences between membranes. The amount of bovine grafting material (%) within the defects significantly decreased from 30 to 90 days. Collagen membranes led to an upregulation of cellular proliferation and adhesion along with increased expression of genes associated with bone healing, particularly the intact fish skin membrane. Despite an increase in the bone formation rate in the defect over time, there was no significant difference between the membranes.