**1. Introduction**

Osseointegration [1] is a central event for oral implant function. This specific bone reaction has been described and studied at length for titanium and other materials. Technical innovations have led to improvements of bone reactions, such as material surface topographical changes [2–4] that have been vastly adopted by the oral implant industry, as well as different forms of chemical surface modulations [5,6]. Such surface related innovations have resulted in improved clinical results and widening of clinical indications [7,8]. However, the specific bone related control mechanisms that lead to osseointegration are still in need of scientific analyses, as are the reasons for marginal bone resorption. Generally speaking, the foreign body reaction (FBR) is accepted as the series of host events that follow the introduction of a material into tissues. The host–biomaterial interaction [9] depends on the type of material, clinical handling and on the tissue where the implant is placed (e.g., bone, skin, and blood vessel), as well as the host specific conditions. The immune system has a central role in the FBR [10–12] where the M1/M2-macrophage phenotype balance has been identified as one of the main controlling factors at the cellular level [13]. Macrophages are thus able to shift between an M1-phenotype (proinflammatory) and an M2-phenotype (reparative/anti-inflammatory), with obvious consequences for tissue reaction to biomaterials, and experimental modulation of this balance has been studied to direct a favorable pathway for bone regeneration [14]. The current authors have demonstrated an early M1/M2 shift around titanium, at 10 days of follow-up towards a dominant M2 macrophage phenotype [15], in contrast to other materials such as polyetheretherketone (PEEK) and Copper (Cu) that present mixed M1/M2 phenotypes at the same short term of follow-up. Osseointegration is thus seen as the result of an FBR which in the long run may achieve a foreign body equilibrium allowing for long term loading of implants [16]. However, the basis for the control of bone metabolism around implants in health and disease remains largely unclear [17]. Particularly the events taking place after the inflammatory period of initial healing and a possible immunological regulation of bone metabolism are examples of important fields for further studies. Our group has demonstrated that titanium activates the immune system when compared to a sham site at 10 and 28 days of follow-up [12]. In Part I of this series of studies (where the current work is Part II), the importance of the specific immune response around different materials when compared to a sham site was demonstrated at an early stage of 10 days [15]. The current study aims at comparing materials that do not osseointegrate, i.e., test materials copper (known to induce a pronounced FBR in soft tissues [18]) and PEEK (considered a bioinert material [19]), to a material that osseointegrates, cp titanium (control) at 10 and 28 days, in order to investigate and compare the respective immune modulation reactions between the inflammatory (10 days) and postinflammatory (28 days) stages of bone healing.

#### **2. Materials and Methods**

The current study consists of an experiment in the rabbit proximal tibia (metaphysis), comparing bone healing on sites where osteotomies were performed and one of three test materials were placed for comparison: titanium (Ti), copper (Cu), or polyether ether ketone (PEEK), where Ti was a control.

All implants were turned with a threaded 0.6 mm pitch height, 3.75 mm width, and Branemark MkIII design (Figure 1). The Ti implants were made of commercially pure titanium grade IV (98.55% Ti, with specified maximum traces of elements Fe, O, N, H, and C for the remaining 1.45%).

**Figure 1.** Implant design with 3.75mm width and 8mm length. Representative image of titanium (Ti) implant; copper (Cu) and polyetheretherketone (PEEK) implants with the same design.

#### *2.1. Surgical Procedure*

This study was performed on 12 mature, female New Zealand White Rabbits (*n* = 6 for each time point, 10 and 28 days, weight 3 to 4 Kg), with the ethical approval from the Ethics Committee for Animal Research (No. 13-011) of the École Nationale Vétérinaire D'Alfors, Maisons-Alfors, Val-de-Marne, France. The 6 animals at 10 days are the same used for Part I of this series of studies [12]. All care was taken to minimize animal pain or discomfort during and after the surgical procedures. For the surgical procedures, the rabbits were placed under general anesthesia using a mixture of medetomidine (Domitor; Zoetis, Florham Park, NJ, USA), ketamine (Imalgène 1000; Merial, Lyon, France), and diazepam (Valium; Roche, Basel, Switzerland) for induction, then applying subcutaneous buprenorphine (Buprecare; Animalcare, York, UK) and intramuscular meloxicam (Metacam; Boehringer Ingelheim Vetmedica, Inc., Ridgefield, CT, USA). A single incision was performed in the internal knee area on each side and the bone exposed for osteotomies and insertion of implants in the sites mentioned above. The surgical site was sutured with a resorbable suture (Vicryl 3/0; Ethicon, Cincinnati, OH, USA) and hemostasis achieved. Following surgery, Fentanyl patches (Duragesic; Janssen Pharmaceutica, Beerse, Belgium) were applied.

The osteotomies were produced with a sequence of increasing diameter twist drills, from 2 mm to 3.15 mm width, and a final countersink bur prepared the cortical part of the bone. The implants used were 3.75 mm in diameter, placed in an underprepared osteotomy to achieve primary (mechanical) stability.

The rabbits were housed in separate cages and were allowed to move and eat freely.

At 10 and 28 days, the rabbits were sacrificed with a lethal injection of sodium pentobarbital (Euthasol; Virbac, Fort Worth, TX, USA). The 6 animals at each time point had the implants removed through unscrewing. On 4 animals at 10 days and 5 animals at 28 days, bone was collected with a 2 mm twist drill from the periphery of the Ti, Cu, and PEEK sites on the most distal portion, and then processed through quantitative-polymerase chain reaction (qPCR). After this, at each time point, the implant sites were removed en bloc for histological processing on the 6 animals.

#### *2.2. Gene Expression Analysis—qPCR*

The bone samples for gene expression analysis at 10 or 28 days were collected from the distal side of the osteotomies of all three groups (following the removal of the implant from the implant sites), with a 2 mm twist drill that removed both cortical and marrow bone in the full length of the osteotomy, to enable the study of the 2 mm peri-implant bone area of each of the Ti, Cu, and PEEK sites. The samples were immediately transferred to separate sterile plastic recipients containing RNA*later* medium (AmbionInc, Austin, TX, USA) for preservation. The samples were then refrigerated first at 4 ◦C and then stored at –20 ◦C until processing.

#### 2.2.1. mRNA Isolation

Samples were homogenized using an ultrasound homogenizer (Sonoplus HD3100, Brandelin) in 1 ml PureZOL and total RNA was isolated via column fractionalization using the AurumTM Total RNA Fatty and Fibrous Tissue Kit (Bio-Rad Laboratories Inc.; Hercules, CA, USA) following the manufacturer's instructions. All the samples were DNAse treated using an on-column DNAse I contained in the kit to remove genomic DNA. The RNA quantity for each sample was analyzed in the NanoDrop 2000 Spectrophotometer (Thermo Scientific; Wilmington, DE, USA). BioRad iScript cDNA synthesis kit (Bio-Rad Laboratories Inc.; Hercules, CA, USA) was then used to convert mRNA into cDNA, following the manufacturer's instructions.

qPCR primers (Tataa Biocenter; Gothenburg, Sweden) were designed following the NCBI Sequence database, including the local factors chosen in order to characterize the immune, inflammatory, and bone metabolic pathways (Tables 1 and 2). All primers had efficiency between 90% and 110%.


#### **Table 1.** Gene sequences.

*NCF-1* (neutrophil cytosolic factor 1); *CD68* (macrosialin); *CD11b* (*MAC-1*, macrophage marker); *CD14* (monocyte differentiation antigen *CD14*); *ARG1* (Arginase 1); *IL-4* (Interleukin 4); *IL-13* (Interleukin 13); *M-CSF* (colony stimulating factor-macrophage); *OPG* (osteoprotegerin); *RANKL* (Receptor activator of nuclear factor kappa-B ligand); *TRAP* (tartrate resistant acid phosphatase); *CathK* (cathepsin K); *PPAR-*γ (peroxisome proliferator activated receptor gamma); *C3* (complement component 3); *C3aR1* (complement component 3a receptor 1); *CD46* (complement regulatory protein); *CD55* (decay accelerating factor for complement); *CD59* (complement regulatory protein); *C5* (complement component 5); *C5aR1* (complement component 5a receptor 1); *CD3* (T cell surface glycoprotein CD3); *CD4* (T cell surface glycoprotein CD4); *CD8* (T cell transmembrane glycoprotein CD8); *CD19* (B-lymphocyte surface protein CD19); *GAPDH* (glyceraldehyde-3-phosphate dehydrogenase); *ACT-*β (actin beta); *LDHA* (lactate dehydrogenase A).

**Table 2.** Correspondence between studied gene expression and biological entities.


*NCF-1* (neutrophil cytosolic factor 1); *CD68* (macrosialin); *CD11b* (*MAC-1*, macrophage marker); *CD14* (monocyte differentiation antigen *CD14*); *ARG1* (Arginase 1); *IL-4* (Interleukin 4); *IL-13* (Interleukin 13); *M-CSF* (colony stimulating factor-macrophage); *OPG* (osteoprotegerin); *RANKL* (Receptor activator of nuclear factor kappa-B ligand); *TRAP* (tartrate resistant acid phosphatase); *CathK* (cathepsin K); *PPAR-*γ (peroxisome proliferator activated receptor gamma); *C3* (complement component 3); *C3aR1* (complement component 3a receptor 1); *CD46* (complement regulatory protein); *CD55* (decay accelerating factor for complement); *CD59* (complement regulatory protein); *C5* (complement component 5); *C5aR1* (complement component 5a receptor 1); *CD3* (T cell surface glycoprotein CD3); *CD4* (T cell surface glycoprotein CD4); *CD8* (T cell transmembrane glycoprotein CD8); *CD19* (B-lymphocyte surface protein CD19); *GAPDH* (glyceraldehyde-3-phosphate dehydrogenase); *ACT-*β (actin beta); *LDHA* (lactate dehydrogenase A).

#### 2.2.2. Amplification Process

Five microliters of SsoAdvanced SYBR™ Green Supermix (Bio-Rad Laboratories Inc.; Hercules, CA, USA) and 1 μL of cDNA template together with 0.4 μM of forward and reverse primer were used

in the qPCR reaction. Each cDNA sample was performed on duplicates. The thermal cycles were performed on the CFX Connect Real-Time System (Bio-Rad Laboratories Inc.; Hercules, CA, USA). The CFX Manager Software 3.0 (Bio-Rad, Hercules, CA, USA) was used for the data analysis.

Three genes (*GAPDH*, *ACT-beta*, and *LDHA*) were selected as reference genes using the geNorm algorithm integrated in the CFX Manager Software. A quantification cycle (Cq) value of the chosen reference genes (Tables 1 and 2) was used as control; hence the mean Cq value of each target gene (Table 1) was normalized against the reference gene's Cq, giving the gene's relative expression. For calculation of fold-change, the ΔΔCq was used, comparing mRNA expressions from the different groups. Significance was set at *p* < 0.05 and the regulation threshold at ×2 fold-change.

#### *2.3. Decalcified Bone Histology*

After removal of the implants from the studied Ti, Cu, and PEEK sites on the 6 subjects of each time point, bone was removed en bloc and preserved in 10% formalin (4% buffered formaldehyde; VWR international, Leuven, Belgium) during 48 h for fixation. Samples were decalcified in Ethylene diamine tetra-acetic acid (10% unbuffered EDTA; Milestone srl, BG, Italy) for 4 weeks, with weekly substitution of the EDTA solution, dehydrated and embedded in paraffin (Tissue-Tek TEC; Sakura Finetek Europe BV, Leiden, NL, USA). Samples were sectioned (4-μm-thick) with a microtome (Microm HM355S; Microm International GmbH, Thermo Fischer Scientific, Walldorf, Germany) and stained with Haematoxylin-Eosin (HE) for histological analysis.
