Biotherapeutic Effect of Gingival Stem Cells Conditioned Medium in Bone Tissue Restoration

Bone tissue engineering is one of the main branches of regenerative medicine. In this field, the use of a scaffold, which supported bone development, in combination with mesenchymal stem cells (MSCs), has promised better outcomes for bone regeneration. In particular, human gingival mesenchymal stem cells (hGMSCs) may present advantages compared to other MSCs, including the easier isolation. However, MSCs’ secretome has attracted much attention for its potential use in tissue regeneration, such as conditioned medium (CM) that contains different soluble factors proved to be useful for the regenerative purposes. In this study, we evaluated the osteogenic capacity of a poly-(lactide) (3D-PLA) scaffold enriched with hGMSCs and hGMSCs derived CM and its ability to regenerate bone defects in rat calvarias. 3D-PLA alone, 3D-PLA + CM or 3D-PLA + hGMSCs with/without CM were implanted in Wistar male rats subjected to calvarial defects. We observed that 3D-PLA scaffold enriched with hGMSCs and CM showed a better osteogenic capacity, being able to repair the calvarial defect as revealed in vivo by morphological evaluation. Moreover, transcriptomic analysis in vitro revealed the upregulation of genes involved in ossification and regulation of ossification in the 3D-PLA + CM + hGMSCs group. All of these results indicate the great osteogenic ability of 3D-PLA + CM + hGMSCs supporting its use in bone regenerative medicine, in particular in the repair of cranial bone defects. Especially, hGMSCs derived CM played a key role in the induction of the osteogenic process and in bone regeneration.


Introduction
Bone is a mineralized connective tissue with important functions, including locomotion, support and protection of soft tissues, calcium and phosphate storage. It is highly dynamic being always resorbed by osteoclasts and neoformed by osteoblasts [1]. In some conditions, bone regeneration is required in large quantity, such as in the case of large bone defects caused by trauma, degeneration or skeletal abnormalities, or when regenerative process is compromised. In this case, bone regeneration represents an important problem and there is a need for treatments. In this context, bone tissue engineering has promised good results.
The use of a scaffold with cells and/or soluble factors has emerged as a promising approach for bone regeneration in the field of regenerative medicine. A biomaterial refers to a matrix that provides and capable of extensive proliferation when maintained in their standard culture conditions; moreover, they are able to differentiate into mesengenic lineages. In hGMSCs cultures, the Alizarin Red staining highlighted the presence of calcium deposition in osteogenic differentiation medium-treated cells after three weeks of induction ( Figure 1C). Oil Red O positive lipid droplets were evident in the cytoplasm of hGMSCs induced to adipogenic commitment after 28 days of induction ( Figure 1D). staining highlighted the presence of calcium deposition in osteogenic differentiation medium-treated cells after three weeks of induction ( Figure 1C). Oil Red O positive lipid droplets were evident in the cytoplasm of hGMSCs induced to adipogenic commitment after 28 days of induction ( Figure 1D).

3D-PLA Scaffold Characterization
The three-dimensional (3D) structural morphology of printed PLA scaffold is characterized by means of scanning electron microscopy (SEM) (Figure 2A). Surface morphology is represented by a modular structure. At high magnification, hGMSCs adhesion onto the substrate is observable. Cells developed a bridge-like structure on the PLA filaments ( Figure 2B).

3D-PLA Scaffold Characterization
The three-dimensional (3D) structural morphology of printed PLA scaffold is characterized by means of scanning electron microscopy (SEM) (Figure 2A). Surface morphology is represented by a modular structure. At high magnification, hGMSCs adhesion onto the substrate is observable. Cells developed a bridge-like structure on the PLA filaments ( Figure 2B). staining highlighted the presence of calcium deposition in osteogenic differentiation medium-treated cells after three weeks of induction ( Figure 1C). Oil Red O positive lipid droplets were evident in the cytoplasm of hGMSCs induced to adipogenic commitment after 28 days of induction ( Figure 1D).

3D-PLA Scaffold Characterization
The three-dimensional (3D) structural morphology of printed PLA scaffold is characterized by means of scanning electron microscopy (SEM) (Figure 2A). Surface morphology is represented by a modular structure. At high magnification, hGMSCs adhesion onto the substrate is observable. Cells developed a bridge-like structure on the PLA filaments ( Figure 2B).

Osteogenic In Vitro Performance
hGMSCs were observed by inverted light microscopy to verify the alizarin red S staining ( Figure 3). Microphotographs showed the red positive staining in 3D-PLA + CM + hGMSCs when compared with 3D-PLA + hGMSCs. Insets in Figure 3B,C are a representative macrophotographs of 3D-PLA placed in multiwell cell culture plates. Obtained results were evaluated using spectrometric analysis after six weeks of culture ( Figure 3D).

Osteogenic In Vitro Performance
hGMSCs were observed by inverted light microscopy to verify the alizarin red S staining ( Figure 3). Microphotographs showed the red positive staining in 3D-PLA + CM + hGMSCs when compared with 3D-PLA + hGMSCs. Insets in Figure 3B,C are a representative macrophotographs of 3D-PLA placed in multiwell cell culture plates. Obtained results were evaluated using spectrometric analysis after six weeks of culture ( Figure 3D). (B) hGMSCs cultured with poly-(lactide) (3D-PLA) scaffold; (C) hGMSCs cultured with 3D-PLA + conditioned medium (CM); (D) histograms of alizarin red S staining quantification. Scale bars represent 10 μm. Graph optical density: ** p < 0.01 was considered statistically significant.

Global Transcriptome Analysis
The transcriptome of hGMSCs, 3D-PLA + hGMSCs and 3D-PLA + CM + hGMSCs was investigated through NGS analysis. We observed that, among the genes that were differentially expressed between these groups, 21 genes were totally inhibited or expressed at very low levels in hGMSCs, while they were expressed in 3D-PLA + hGMSCs and/or 3D-PLA + CM + hGMSCs. Interestingly, the gene ontology (GO) analysis of these genes revealed that eight genes belonged to the category ossification, eight genes to the category regulation of ossification, while five genes were in common for both categories. The analysis showed that the genes BCAP29 and BMP2K were slightly downregulated in 3D-PLA + hGMSCs compared to hGMSCs. Instead, these two genes were upregulated in 3D-PLA + CM + hGMSCs compared to the other groups. The genes DHRS3, FAM20C, FHL2 and TOB2 were upregulated in 3D-PLA + hGMSCs compared to hGMSCs ( Figure 4). In addition, the expression levels of these genes were upregulated in 3D-PLA + CM + hGMSCs compared to 3D-PLA + hGMSCs ( Figure 4). Instead, the genes ASF1A, GDF5, HDAC7, ID3, INTU,

Global Transcriptome Analysis
The transcriptome of hGMSCs, 3D-PLA + hGMSCs and 3D-PLA + CM + hGMSCs was investigated through NGS analysis. We observed that, among the genes that were differentially expressed between these groups, 21 genes were totally inhibited or expressed at very low levels in hGMSCs, while they were expressed in 3D-PLA + hGMSCs and/or 3D-PLA + CM + hGMSCs. Interestingly, the gene ontology (GO) analysis of these genes revealed that eight genes belonged to the category ossification, eight genes to the category regulation of ossification, while five genes were in common for both categories. The analysis showed that the genes BCAP29 and BMP2K were slightly downregulated in 3D-PLA + hGMSCs compared to hGMSCs. Instead, these two genes were upregulated in 3D-PLA + CM + hGMSCs compared to the other groups. The genes DHRS3, FAM20C, FHL2 and TOB2 were upregulated in 3D-PLA + hGMSCs compared to hGMSCs ( Figure 4). In addition, the expression levels of these genes were upregulated in 3D-PLA + CM + hGMSCs compared to 3D-PLA + hGMSCs ( Figure 4)  Gene expression levels, fold changes, gene ontology (GO) processes and false discovery rate (FDR) were reported in Table 1. Gene expression levels, fold changes, gene ontology (GO) processes and false discovery rate (FDR) were reported in Table 1. Table 1. Gene expression levels, fold changes, gene ontology (GO) and false discovery rate (FDR) of differentially expressed genes in human gingival mesenchymal stem cells (hGMSCs), poly-(lactide) (3D-PLA) scaffold + hGMSCs and 3D-PLA + conditioned medium (CM) + hGMSCs.

3D-PLA In Vivo Evaluation
Bone regeneration in the implanted areas was evaluated at six weeks after surgery. Macroscopic evaluation showed a better regeneration capacity of 3D-PLA + CM + hGMSCs.
In the 3D-PLA scaffold group, the growing woven extracellular matrix (ECM) not fully covered the interstitial area of the biomaterial porous at six weeks after surgery ( Figure 5A). At high magnification, empty areas are visible inside the biomaterial structure ( Figure 5B). The 3D-PLA + hGMSCs group showed more ECM deposition inside the biomaterial scaffold when compared to the 3D-PLA group ( Figure 5C). At high magnification ECM, without signs of mineralization, is present on the filaments side and at the interface of the native bone and scaffold ( Figure 5D).

3D-PLA In Vivo Evaluation
Bone regeneration in the implanted areas was evaluated at six weeks after surgery. Macroscopic evaluation showed a better regeneration capacity of 3D-PLA + CM + hGMSCs.
In the 3D-PLA scaffold group, the growing woven extracellular matrix (ECM) not fully covered the interstitial area of the biomaterial porous at six weeks after surgery ( Figure 5A). At high magnification, empty areas are visible inside the biomaterial structure ( Figure 5B). The 3D-PLA + hGMSCs group showed more ECM deposition inside the biomaterial scaffold when compared to the 3D-PLA group ( Figure 5C). At high magnification ECM, without signs of mineralization, is present on the filaments side and at the interface of the native bone and scaffold ( Figure 5D). In the 3D-PLA + CM group, a strong ECM deposition inside the biomaterial structure and on the native bone tissue was visible ( Figure 6A). At high magnification, different new blood vessel structures are visible in the ECM and on the interface of scaffold and host tissue, in order to indicate the beginning of the regeneration process ( Figure 6B). In contrast, less blood vessels are visible in 3D-PLA + CM + hGMSCs ( Figure 6C,D).
A summary of these results is reported in Table 2. In the 3D-PLA + CM group, a strong ECM deposition inside the biomaterial structure and on the native bone tissue was visible ( Figure 6A). At high magnification, different new blood vessel structures are visible in the ECM and on the interface of scaffold and host tissue, in order to indicate the beginning of the regeneration process ( Figure 6B). In contrast, less blood vessels are visible in 3D-PLA + CM + hGMSCs ( Figure 6C,D).
A summary of these results is reported in Table 2.

3D-PLA (N = 4)
Rats subjected to scraping of the cortical calvaria bone tissue and implant of 3D-PLA Evaluation 6 weeks after surgery evidenced that ECM not fully covered the interstitial area of the biomaterial porous and empty areas were present inside the biomaterial structure.

Discussion
Materials for tissue engineering should possess specific features in order to be considered as ideal material for grafting in bone tissue. Mainly, they should be biocompatible, bioactive, and act as a 3D template for cellular activity in order to be well integrated with the surrounding biological environment of the implant site and to trigger specific cellular responses [24]. 3D printing is becoming an increasingly common technique to fabricate scaffolds and devices for tissue engineering applications that were previously difficult to build [25]. A 3D-printing technique for medical devices has become a popular tool to develop scaffolds for clinical use, and they can be customized, fabricated feasibly and economically advantageous [26]. PLA is an absorbable polymer that has often been used in skeletal tissue engineering that ensure mechanical stability while degrading. The degradation products of this polymer are non-toxic and removed by natural metabolic pathways [27].
In our study, thick PLA scaffolds with square pores were produced by a custom-made 3D printer based on this principle. The printed scaffolds were then functionalized with hGMSCs and/or CM and then evaluated for their capacity to induce in vitro an osteogenic process and in vivo to start the osteogenic repair of bone defects when grafted in rat calvaria.
Actually, MSCs have been applied for autologous therapy in combination with platelet-rich plasma and/or scaffolds in distraction osteogenesis. In particular, therapeutic ability of MSCs has been attributed at least to three major mechanisms: (1) homing process, whereby the systemic stem cell delivery results in cell migration to specific areas of acute injury via chemical gradients; (2) differentiating process, in order to locally regenerate or replace the damaged tissues; and (3) secreting bioactive molecules, in order to affect local and systemic processes via paracrine mechanisms [28,29].
Human GMSCs have recently been proposed as a novel option in regenerative therapy. They express MSC surface markers, such as Oct3/4, Sox-2, SSEA-4, CD29, CD44, CD73, CD90 and CD105, and lacking the expression for CD34, CD14 and CD45 [30]. Moreover, they are able to differentiate into adipogenic and osteogenic lineage as demonstrated by biochemical analysis and possess a canonical fibroblast like morphology. hGMSCs are thought to be good candidate cell sources, easy to access and manipulate, in order to use in tissue engineering [31]. A recent study reported that the oral MSCs cultured onto 3D-scaffolds grafted in the mouse calvaria lead an osteointegration and vascularisation process.
Previously, we have demonstrated that the PLA scaffolds are non-cytotoxic and biocompatible. In our experiments, SEM analysis revealed that hGMSCs are able to adhere and grown onto the scaffold, covering the biomaterial surface without any morphological changes in cytoskeleton rearrangement. Indeed, PLA showed a good biocompatibility with another MSC type derived from oral tissues and the stem cells from human exfoliated deciduous teeth (SHED), inducing also the mineralization of SHED [32]. In addition, a star-shaped PLA was reported to induce the differentiation of human adipose MSCs toward the osteogenic lineage in vitro and enhanced the mineralized tissue formation in vivo [33]. These features, in accordance with our results, indicated that PLA is a good biomaterial for bone tissue engineering.
To improve the performance of the PLA scaffold in our study, we evaluated the effects of the combination of the scaffold with the CM and hGMSCs. MSCs possess immunomodulatory and anti-inflammatory properties and using paracrine signaling, they can exert their action in the implanted site and favorably affect in vitro the extracellular matrix remodeling [34,35]. The secretome of MSCs contains many growth factors, such as insulin growth factor that can affect the migration, angiogenesis, and osteogenic differentiation and possess a great potential effect for bone and periodontal tissue regeneration [35]. CM derived from human adult MSCs has been already applied to accelerate bone formation in preclinical animal models [36]. Osugi and co-workers [19] showed that CM derived from human bone marrow-MSCs, containing different cytokines including insulin-like growth factor-1 and vascular endothelial growth factor, enhanced the migration, proliferation, and expression of osteogenic marker genes in rat MSCs in vitro. In addition, CM increased bone regeneration in vivo in rats with calvaria defects, and, in particular, after eight weeks, the new regenerated bone nearly covered the defect. In particular, it seems that CM improved bone regeneration through the mobilization of endogenous MSCs [19].
The results from our study suggest that the 3D-PLA + CM + hGMSCs might improve the osteogenic process in vitro as demonstrated by alizarin red staining; indeed, calcium depositions are more evident in the presence of CM when compared with the other experimental groups. Thus, the utilization of the CM not only might halt bone loss in diseases such as periodontitis, but also could improve the quality of the new bone with increased mineralization and connectivity.
The transcriptome profile showed that 21 genes were inhibited or expressed at low levels in hGMSCs, but expressed in 3D-PLA + hGMSCs and/or 3D-PLA + CM + hGMSCs. hGMSCs expressed only the genes BCAP29 and BMP2K at very low levels. 3D-PLA + hGMSCs showed a slight downregulation of these two genes, while the other four genes were expressed and upregulated compared to hGMSCs. All the 21 genes were expressed in 3D-PLA + CM + hGMSCs. In particular, the expression of all these genes was upregulated in 3D-PLA + CM + hGMSCs compared to the other groups. Interestingly, GO analysis revealed that these genes belonged to the categories involved in ossification and regulation of ossification. Then, NGS analysis showed that the hGMSCs grown on the 3D-PLA scaffold enriched with the CM expressed osteogenic genes compared to hGMSCs and 3D-PLA + hGMSCs, suggesting that the CM plays a main role in the induction of an osteogenic transcriptional program.
In earlier in vivo studies, 3D-PLA has demonstrated a well integration with the host tissue. Our in vivo results confirmed the main role played by CM in bone regeneration. Indeed, in the 3D-PLA group, ECM not fully covered the interstitial area of the biomaterial, while the 3D-PLA + hGMSCs group showed more ECM deposition inside the biomaterial scaffold but without signs of mineralization. Instead, the histological analysis of bone regeneration in the calvarial defects revealed improvement of new bone formation in the group implanted with 3D-PLA + CM as reflected by the increase in ECM formation and bone contact when compared to the other groups by the six weeks time-point. Moreover, the macroscopic evaluation showed a better bone repair in 3D-PLA + CM + hGMSCs group.
We can suggest that the implant of 3D-PLA enriched with both CM and hGMSCs may present some advantages. First, cytokines and growth factors contained in the CM could activate an osteogenic differentiation program in hGMSCs, increasing osteogenic gene expression and mineralization as demonstrated by our in vitro results. Given that it has already been demonstrated that GMSCs and CM participated in bone repair, also recruiting bone progenitor cells and endogenous MSCs [13,19], we can suggest that in vivo CM may act inducing the mobilization and then the osteogenic differentiation of both hGMSCs and endogenous MSCs. Second, a problem of stem cell therapy is the poor viability of cells after the implant. However, the presence of the scaffold and the CM could increase the viability rate of hGMSCs. Third, hGMSCs in situ may continue to release factors useful for the regenerative process.
The porous 3D-PLA scaffold can promote the exchange of body fluid and vascularization, but the CM could have a key role in the osteoblasts recruiting in the surrounding tissues stimulating the synthesis and accumulation of ECM, leading to an enhanced osteoid deposition [37]. Our data are therefore in agreement with the promising results obtained on the use of CM that accelerates the formation of new bone callus stimulating the recruitment of endogenous bone marrow derived-MSCs [36]. To our knowledge, this is the first study to evaluate the effects of CM derived from hGMSCs in bone tissue regeneration. Given that hGMSCs present advantages compared to other MSCs, such as easy isolation, the fact that their CM shows efficacy in tissue regeneration, encourage CM deep characterization in order to allow its use in the clinical practice in the future.

Cell Culture and Characterization
Written approval for gingival biopsy collection was obtained from the Medical Ethics Committee at the Medical School, "G. d'Annunzio" University, Chieti, Italy and each participant gave informed consent. Gingival tissue biopsies were obtained from healthy adult volunteers with no gingival inflammation as previously described by Diomede et al. [38]. To define the surface molecules flow citometry was performed as previously reported [39]. Plastic adherent cells were stained with toluidine blue solution and to evaluate cell morphology were observed at light microscopy (Leica, DMIL, Milan, Italy) [40]. To evaluate the capacity to differentiate into mesengenic lineages, cells were cultured under specific culture conditions, osteogenic and adipogenic respectively as reported by Ballerini et al. [41].

Scaffold Development and Three-Dimensional Scaffold Constructs
The scaffold was obtained from a commercial poly-(lactide) (PLA) (Kaytech srl, Ancona, Italy). Samples were developed with a commercial CAD software (Rhinoceros 5, McNeel Europe, Barcelona, Spain) as previously described; the projects were then applied to a printing slicing software (Cura 15.04, Ultimaker B.V., Geldermalsen, The Netherlands). The sliced project was finally transferred to a commercial fuse filament fabrication 3D printer (DeltaWASP 2040; CSP srl, Massa Lombarda, Italy) [42].

Scanning Electron Microscopy (SEM) Characterization
A Scanning Electron Microscope (EVO 50 XVP; Zeiss, Jena, Germany) was used to image the surfaces of the scaffold; the specimens were sputtered with 4-8 nm of gold and then mounted on carbon tape dots [43].

Conditioned Medium (CM) Collection
The CM, after 72 h of incubation, was collected from 15 × 10 3 /cm 2 hGMSCs at the 2nd passage. The CM was centrifuged at 1200 rpm for 5 min to eliminate suspension cells and debris. The supernatants were recentrifuged at 3000 rpm for 3 min, followed by collection of the secondary supernatants. Subsequently, 1 mL of secondary supernatants was resuspended in 3 mL of ice aceton and maintained over night at 4 • C, and after centrifuged at 16.000 rpm for 12 min at 4 • C (Centrifuge 5804 R, Eppendorf, Milan, Italy). The suspension was lysated in radioimmunoprecipitation assay (RIPA) buffer and quantified by means Breadford assay [34,44]. Total proteins obtained were 125 µg/µL.

In Vitro Osteogenesis Performance
hGMSCs were cultured at 8 × 10 3 cells/cm 2 in standard culture medium (MSCGM-CD) (Lonza, Basilea, Switzerland). 3D-PLA was pre-treated for 24 h under agitation (MacsMix, Milthenyi, Bologna, Italy) with 5 mL of CM. Evaluation of calcium deposition and ECM mineralization was obtained by Alizarin Red S (ARS) staining assay performed after 6 weeks. Cells were washed with PBS, fixed in 10% (v/v) formaldehyde (Sigma-Aldrich, Milan, Italy) for 30 min and washed twice with abundant dH 2 O prior to addition 0.5% Alizarin red S in H 2 O, pH 4.0, for 1 h at room temperature. After cell incubation under gentle shaking, cells were washed with dH 2 O four times for 5 min. For staining quantification, 800 µL 10% (v/v) acetic acid was added to each well. Cells incubated for 30 min were scraped from the plate, transferred into a 1.5 mL vial and vortexed for 30 s. The obtained suspension, overlaid with 500 µL mineral oil (Sigma-Aldrich), was heated to 85 • C for 10 min, then transferred to ice for 5 min, carefully avoiding opening of the tubes until fully cooled, and centrifuged at 20,000× g for 15 min. In addition, 500 µL of the supernatant were placed into a new 1.5 mL vial and 200 µL of 10% (v/v) ammonium hydroxide was added (pH 4.1-pH 4.5).
Furthermore, 150 µL of the supernatant obtained from cultures were read in triplicate at 405 nm by a spectrophotometer (Synergy HT, BioTek, Bad Friedrichshall, Germany).

RNA Sequencing and Library Preparation
RNA sequencing libraries were obtained using the TruSeq RNA Access library kit (Illumina, Inc., San Diego, CA, USA) following the manufacturer's instructions. In summary, RNA (50 ng) of each sample was fragmented at 94 • C for 8 min. The synthesis of first strand cDNA was carried out using random hexamers and the SuperScript II Reverse Transcriptase (Invitrogen, Milan, Italy) at 25 • C for 10 min, 42 • C for 15 min, and 70 • C for 15 min.
In the second strand cDNA synthesis, the RNA templates were removed and a second replacement strand was synthesized by dUTP internalization, generating a double strands cDNA. In order to purify the blunt-ended-cDNA, AMPure XP beads (Beckman Coulter, Brea, CA, USA) were used. Afterwards, 3 ends of the cDNA were adenylated to allow the adaptor ligation in the following step. After adaptor ligation, the libraries were purified with AMPure XP beads and a first PCR amplification step was carried out in order to enrich those DNA fragments with adaptors on both ends but also to increase the amount of DNA in the library (15 cycles of 98 • C for 10 s, 60 • C for 30 s, and 72 • C for 30 s). After library validation, 200 ng of each DNA library was combined and the first hybridization step was performed using exome capture probes according to a standardized protocol at the following conditions: 18 cycles of 1 min of incubation, starting at 94 • C, and then decreasing by 2 • C for every cycle. Magnetic beads coated with streptavidin were used to capture probes hybridized to the target regions. Then, the enriched libraries were eluted from the beads and a second cycle of hybridization was performed to obtain a wide specificity of regions of capture. Afterwards, a second capture step using streptavidin-coated beads was performed and two heated washes were done in order to remove nonspecific binding from the beads. After that, the libraries were purified through the AMPure XP bead, and amplified following the protocol (10 cycles; incubation at 98 • C for 10 s, incubation at 60 • C for 30 s and incubation at 72 • C for 30 s), followed by a purification step. Libraries were quantified by qPCR using KAPA Library Quantification Kit-Illumina/ABI Prism (Kapa Biosystems, Inc., Wilmington, MA, USA) and validated with the Agilent High Sensitivity Kit on a Bioanalyzer (Cernusco sul Naviglio, Italy). The size of the DNA fragments has been set in a range of 200-650 bp and peaked around 250 bp. Libraries were normalized to 16 pM and subjected to cluster, and single read sequencing was executed for 300 cycles on a MiSeq instrument (Illumina), according to the protocol guidelines. The produced libraries were loaded for clustering on a MiSeq Flow Cell v2 and sequenced with a MiSeq Instrument (Illumina). The validation of the cluster density was performed by the software of the instrument throughout the run.

NGS Data Processing
Data obtained by NGS analysis were analysed. In particular, the sequence reads were subjected to the demultiplexing process in order to have a separation of the sequence reads in different files for each index tag/sample by using the CASAVA algorithm (CASAVA, version 1.8.2; Illumina, Inc., San Diego, CA, USA). In order to perform the alignment of the sequences, the RNA-Seq Alignment version 1.0.0 (Illumina) and the reference sequence "Homo sapiens University of California, Santa Cruz (UCSC) hg19" were used. For the Read mapping, the TopHat 2 (Bowtie 1) was used. The fragments per kilobase of exon per million fragments mapped (FPKM) values were calculated for each sample using the normalized read counts for each annotated gene: ([1000 × read count] ÷ [number of gene covered bases × number of mapped fragments in million]). Unmapped reads were deleted, preserving only read pairs with both reads aligned to the reference sequence "Homo sapiens UCSC hg19". The comparison between two different samples was carried out by a scatter plot of the LOG2 of the FPKM.

Statistical Analysis
Statistical analysis was accomplished using ANOVA and Tukey's post hoc analysis (p < 0.05). The statistical analysis on the read counts was carried out with the Cufflinks Assembly and DE package version 2.0.0 to establish the proportion of differentially expressed genes for a q-value (FDR) < 0.05. The GO analysis of the genes differentially expressed between experimental group were performed by free tools "Gene Ontolology Consortiun" (available online at http://www.geneontology.org/).

Animals
Male Wistar rats weighing 300-350 g were used for this experiment. Animals were acquired from Harlan, Milan, Italy and housed in individually ventilated cages and maintained under 12 h light/dark cycles, at 21 ± 1 • C and 50-55% humidity with food and water ad libitum.

Ethics Statement for Animal Use
All animal care and use was accomplished according to the European Organization Guidelines for Animal Welfare. The study has been authorized by the Ministry of Health "General Direction of animal health and veterinary drug" (Authorization 768/2016-PR 28/07/2016-D.lgs 26/2014). The experiments were planned in such a way to minimize the total number of rats needed for the study.

Scaffold Grafting
To implant the scaffold, rats were first anesthetized with a combination of tiletamine and xylazine (10 mL/kg, intraperitoneal; i.p.). Afterwards, the implant site was prepared with iodopovinone (Betadine) after trichotomy. Following a median sagittal incision of about 1.0 cm in the frontoparietal region, a total thickness cut was applied; the calvary was then exposed in the frontal area and in the parietal areas. The circular section bone receiving site, with a diameter of 5 mm and a height of 0.25 mm, was injured by means of a rotary instrument at a controlled speed (trephine milling machine, Alpha Bio-Tec, HTD Consulting S.r.l., Siena, Italy) under constant irrigation of a physiological solution.
For their texture and flexibility, the following 3D-PLA, 3D-PLA + hGMSCs, 3D-PLA + CM, and 3D-PLA + CM + hGMSCs were easily inserted in contact with bone tissue to cover the damaged area. The skin flap was then sutured with Caprosyn 6-0 synthetic monofilament absorbable sutures (Covidien AG, Neuhausen am Rheinfall, Switzerland), using interrupted points. Standard feeding and hydration were maintained constant throughout the post-operative phase.

Experimental Design
Rats were randomly distributed into the following groups (N = 16 total animals): After 6 weeks, the animals were euthanized by intravenous administration of Tanax (5 mL/kg body weight) and their calvariae were processed for morphological analysis.
The specimens were fixed for 72 h in 10% formalin solution, dehydrated in ascending graded alcohols and embedded in LR White resin (Sigma-Aldrich). After polymerization, undecalcified oriented cut sections of 50 µm were prepared and ground down to about 30 µm by using the TT System (TMA2, Grottammare, Italy).
The sections before to be stained, were analysed with the CLSM (LSM510 META, Zeiss) while, after double staining procedure with methylene blue and fuchsine acid solutions they were observed under a light microscope.
The investigation was carried out by means of a bright-field light microscope (Leica Microsystem) connected to a high-resolution digital camera DFC425B Leica (Leica Microsystem) [45]. Three-dimensional reconstruction has been obtained by means ZEN2 software (Zeiss). Data and statistical analysis were performed by means the Statistical Package for Social Science (SPSS, v.21.0, IBM Analytics, Armonk, NY, USA).

Conclusions
Overall, our data show that the 3D-PLA + CM + hGMSCs induced the formation of new bone and the osseointegration process in the grafted site. Our results confirm previous studies that indicated the utility of the PLA scaffold for bone regeneration. In addition, both in vitro and in vivo results indicated that CM produced by hGMSCs plays a key role in the osteogenic process and in bone regeneration. This is an important result because the collection of hGMSCs is easier compared to other MSCs. Then, CM production from these cells is also easier and less expensive and may cause less problems to the patients. We believe that our findings encourage the use of biofunctionalized scaffold and CM characterization to promote bone regeneration.