Osteogenic Differentiation Modulates the Cytokine, Chemokine, and Growth Factor Profile of ASCs and SHED

Great efforts have been made to improve bone regeneration techniques owing to a growing variety of sources of stem cells suitable for autologous transplants. Specifically, adipose-derived stem cells (ASCs) and stems cells from human exfoliated deciduous teeth (SHED) hold great potential for bone tissue engineering and cell therapy. After a preliminary characterization of the main biomolecules ASCs and SHED released in their conditioned media, cells were kept both in normal and osteo-inducing conditions. Conventional assays were performed to prove their osteogenic potential such as quantitative real-time polymerase chain reaction (qRT-PCR) (for RUNX-2, collagen type I, osteopontin and osteonectin), alkaline phosphatase activity, osteocalcin production, and von Kossa staining. Conditioned media were tested again after the osteogenic induction and compared to maintaining condition both at base line and after 14 days of culture. The osteogenic condition inhibited the release of all the biomolecules, with the exception, concerning SHED, of growth-regulated alpha protein precursor (GROα), and, to a lesser extent, interleukin (IL)-8. In conclusion, our data support that undifferentiated ASCs and SHED may be preferable to committed ones for general cell therapy approaches, due to their higher paracrine activity. Osteoinduction significantly affects the cytokine, chemokine, and growth factor profile in a differential way, as SHED kept a more pronounced pro-angiogenic signature than ASCs.


Introduction
With more than 2 million bone grafts performed annually worldwide, bone reconstruction is a primary task of regenerative medicine [1]. Autologous bone graft is deemed the safest and most effective grafting procedure [2]. However, the autograft entails a surgical "donor" site, which often brings additional morbidity, including pain and infections [3]. Moreover, the bone source is usually

ASCs and SHED Displayed Different Cell Morphology and Bimolecular Profiles at First Passage
At the first in vitro cell culture passage after harvesting, ASCs appeared more spread out ( Figure 1A), while SHED presented a higher nucleus to cytoplasm ratio ( Figure 1B). ASCs and SHED showed a comparable phenotype consistent with that of MSCs [27]. Both cell types expressed the main mesenchymal markers CD105, CD44, CD73, and CD90 and were negative for CD45. No significant difference in the percentage expression of these markers was found in the two populations ( Figure 1C,D).

Cell Viability
The growth of ASCs and SHED was estimated under maintenance (growth medium, GM) and osteogenic (osteogenic medium, OM) conditions at day 1, 3, and 7 ( Figure 3). Starting from day 1, the proliferation rate was prevented in both ASCs and SHED grown in OM compared to GM. This trend became statistically significant at days 3 and 7 ( Figure 3).

Cell Viability
The growth of ASCs and SHED was estimated under maintenance (growth medium, GM) and osteogenic (osteogenic medium, OM) conditions at day 1, 3, and 7 ( Figure 3). Starting from day 1, the proliferation rate was prevented in both ASCs and SHED grown in OM compared to GM. This trend became statistically significant at days 3 and 7 ( Figure 3). Figure 3. Cell viability assay. The assay was performed on ASCs and SHED cultured in growth medium (GM) or osteogenic medium (OM). Evaluations were made at 1, 3, 7, and 14 days from the beginning of the differentiation using CellTiter-Glo ® luminescent assay. Cell number is expressed as relative luminescence units (RLU). The symbol (*) indicates a significant difference between basal condition (GM) and osteogenic media (OM) considering a p-value < 0.05.

Early Osteogenic Differentiation of ASCs and SHED
To assess the early osteogenic differentiation of ASCs and SHED, the transcript levels of RUNX-2 ( Figure 4A) and collagen type I ( Figure 4B) genes were evaluated through quantitative RT-PCR at days 3, 7, and 14. As a complementary assay to monitor the post-transcriptional level, alkaline phosphatase activity was also tested at day 7 ( Figure 4C). These forms of evidence supported that both ASCs and SHED underwent proper osteogenic commitment, the bone markers being constantly higher in osteodifferentiated cells compared to the control ( Figure 4). The statistical significance was achieved at day 7 in every assay. . Cell viability assay. The assay was performed on ASCs and SHED cultured in growth medium (GM) or osteogenic medium (OM). Evaluations were made at 1, 3, 7, and 14 days from the beginning of the differentiation using CellTiter-Glo ® luminescent assay. Cell number is expressed as relative luminescence units (RLU). The symbol (*) indicates a significant difference between basal condition (GM) and osteogenic media (OM) considering a p-value < 0.05.

Early Osteogenic Differentiation of ASCs and SHED
To assess the early osteogenic differentiation of ASCs and SHED, the transcript levels of RUNX-2 ( Figure 4A) and collagen type I ( Figure 4B) genes were evaluated through quantitative RT-PCR at days 3, 7, and 14. As a complementary assay to monitor the post-transcriptional level, alkaline phosphatase activity was also tested at day 7 ( Figure 4C). These forms of evidence supported that both ASCs and SHED underwent proper osteogenic commitment, the bone markers being constantly higher in osteodifferentiated cells compared to the control ( Figure 4). The statistical significance was achieved at day 7 in every assay.

Late Osteogenic Differentiation of ASCs and SHED
Late osteogenic differentiation of ASCs and SHED was evaluated by quantifying the transcriptional levels of osteopontin ( Figure 5A) and osteonectin ( Figure 5B) genes, which resulted in more expressed in OM condition at both day 7 and 14. Likewise, a significant enhancement of released osteocalcin was observed in osteoinduced cells compared to the controls at day 21 ( Figure 5C). All these data clearly showed that ASCs and SHED were properly osteodifferentiated. As a further confirmation, the presence of calcified matrix was proven by von Kossa staining in osteoinduced ASCs and SHED at 21 days (respectively Figure 5D,E).

Late Osteogenic Differentiation of ASCs and SHED
Late osteogenic differentiation of ASCs and SHED was evaluated by quantifying the transcriptional levels of osteopontin ( Figure 5A) and osteonectin ( Figure 5B) genes, which resulted in more expressed in OM condition at both day 7 and 14. Likewise, a significant enhancement of released osteocalcin was observed in osteoinduced cells compared to the controls at day 21 ( Figure  5C). All these data clearly showed that ASCs and SHED were properly osteodifferentiated. As a further confirmation, the presence of calcified matrix was proven by von Kossa staining in osteoinduced ASCs and SHED at 21 days (respectively Figure 5D,E).

The Biomolecular Profile of ASCs and SHED Was Modulated upon Culture in OM
The expression profile of growth factors, cytokines, and interleukins produced by ASCs and SHED was again detected with the same panel that was evaluated after the harvest ( Figure 2). After 14 days of culturing in GM ( Figure 6A,B, Table 2), ASCs increased their levels of IL-6, IL-8, and MCP, while SHED released more IL-6, IL-8, GROα, MIF, and SDF1α. The comparison of the cells conditioned in OM for 14 days ( Figure 6C,D, Table 2) with the baseline of freshly harvested cells revealed that the osteogenic condition inhibits the release of all the biomolecules expressed by ASCs and SHED, with the exception for the latter of GROα, and, to a lesser extent, IL-8.

The Biomolecular Profile of ASCs and SHED Was Modulated upon Culture in OM
The expression profile of growth factors, cytokines, and interleukins produced by ASCs and SHED was again detected with the same panel that was evaluated after the harvest (Figure 2). After 14 days of culturing in GM ( Figure 6A,B, Table 2), ASCs increased their levels of IL-6, IL-8, and MCP, while SHED released more IL-6, IL-8, GROα, MIF, and SDF1α. The comparison of the cells conditioned in OM for 14 days ( Figure 6C,D, Table 2) with the baseline of freshly harvested cells revealed that the osteogenic condition inhibits the release of all the biomolecules expressed by ASCs and SHED, with the exception for the latter of GROα, and, to a lesser extent, IL-8.     Table 2. Bio-Plex quantification of factors produced by ASCs and SHED after 14 days of culturing in GM and upon osteoinduction in OM. The evaluated factors are expressed in pg/mL and are: interleukin-6 (IL-6), interleukin-8 (IL-8), interleukin-10 (IL-10), interleukin-12 (IL-12), monocyte chemoattractant protein-1 (MCP-1), vascular endothelial growth factor (VEGF), growth-regulated alpha protein precursor (GROα), hepatocyte growth factor (HGF), macrophage migration inhibitory factor (MIF), β-nerve growth factor (β-NGF), stem cell factor (SCF), and stromal cell-derived factor 1-α (SDF-1α). Italic fonts of numbers indicate a significant difference between the two stem cell populations, considering a p-value < 0.05.

Discussion
In recent years, great efforts have been made to improve bone regeneration techniques, owing not only to the technological innovations in the field of material science, but also to a growing variety of sources of stem cells for autologous transplants [8,28]. Among all the possible options, MSCs hold great potential for bone tissue engineering [29]. In the present work, we focused our attention on ASCs and SHED [17,20,21,30] respectively due to their abundance and their accepted bankability based on cryopreservation [31]. Interestingly, we showed that ASCs and SHED display distinctive morphologies when they adhere on plastic culture dishes, as it was portrayed even here ( Figure 1A,B) through immunofluorescent staining. The standard panel of markers used [27] qualified both cell types as mesenchymal stem cells ( Figure 1C,D).
Despite their common mesenchymal nature, ASCs and SHED derive from different tissues, and the microenvironment in which these MSCs live may deeply affect their gene expression profile [32] as well as their secretome. Thus, the purpose of this study was to evaluate and compare the effect of osteodifferentiation on the biomolecular expression profile of ASCs and SHED. To accomplish this task, a standard panel of growth factors, cytokines, and interleukins was evaluated (Figure 2), through Bio-Plex assay [25]. This technique has multiple advantages over common ELISA kits, since it permits a large number of molecules to be analyzed in the same sample and also allows reproducible and comparable titration curves to be established [33,34].
We found that ASCs and SHED strongly differed in their basal expression of several biomolecules. Important factors like HGF and SDF-1α were abundant in SHED but barely detectable in ASCs. Cloned as a mitogenic protein for hepatocytes [35], HGF promotes morphogenesis and cell survival. HGF was also proven to induce epithelial tubulogenesis/morphogenesis [36] and it has become a promising candidate for treating patients with impaired tissue function [37]. As for SDF-1α, it is a strong chemoattractant for stem cells that has received growing attention as a means to recruit endogenous progenitor cells possibly leading to regeneration in situ [38]. SCF, the ligand of the tyrosine kinase receptor c-KIT, and β-NGF were more expressed by SHED than ASCs, but did not reach high levels. These biomolecules might sustain trophic mechanisms and cooperate to allow for the successful use of MSCs for musculoskeletal applications [39]. The pro-inflammatory cytokine MIF was almost equally produced in both cell types. ASCs released higher levels of multifunctional interleukins (IL-6, IL-8, IL-10, IL-12) and VEGF, the prototypical pro-angiogenic factor, than SHED, whilst GROα was scarce in SHED, but fairly represented in ASCs.
Surprisingly, very little information is available in the literature concerning secretome of ASCs and SHED; indeed, it is quite rare to find reports where more than five cytokines/growth factors are analyzed at the same time. Different works reported the release of high amounts of VEGF by ASCs [40][41][42], similar to our result. Park et al. [43] examined the ASC secretion profile and its potential usage in treating skin ageing. Moreover, through the usage of enzyme-linked immunosorbent assay kits, these authors quantified HGF, which, in discord with our observation, resulted in the range of 670 pg/mL. The conspicuous production of HGF in SHED was instead described by Yamaguchi S et al. as was the low amount released by ASCs used in comparison [26], results which are pretty consistently in line with the outcome of our study (Figure 2).
Osteogenic condition inhibited cell proliferation, as it could be easily predicted, in accordance with previous literature [44,45]. Likewise, the efficacy of the osteo-inductive condition adopted was assessed with standard methods, which consistently confirmed that the in vitro differentiation sought occurred properly. Interestingly, we found that ASCs increased their levels of IL-6, IL-8, and MCP, while SHED augmented the release of IL-6, IL-8, GROα, MIF, and SDF1α when cells became confluent, as it was clearly portrayed in the Bio-Plex analysis at 14 days ( Figure 6A).
The secretion of paracrine factors has been pointed out as a major way to enhance the healing process in numerous approaches based on ASCs [46,47] and SHED [48,49]. Based on these forms of evidence, the effect elicited by osteoinduction on the biomolecular profile of ASCs and SHED becomes of great interest. Osteogenic commitment tended to considerably downregulate the paracrine activity and thus the release of notably useful factors when compared to the baseline condition (shortly after cell harvest). Quite relevantly, all the biomolecules decreased, with the only exception of GROα and IL-8 in osteoinduced SHED. GROα, also known as CXCL1, was described among the key angiogenic chemokines [50], while the pro-angiogenic role of IL-8 has been acknowledged since the paramount study by Koch and colleagues [51]. Owing to this pro-angiogenic cue, committed SHED could prove useful to implement ex vivo bone regenerative Good Manufacturing Practice compliant protocols, especially in the dental field.

Primary Cell Harvest and Culture
The primary human cells used in this study, adipose-derived stem cells (ASCs) and stem cells from human exfoliated eciduous teeth (SHED), were obtained from human volunteers between 1 January and 30 June 2013. The research was conducted following the protocols approved (in October 2012) by the Ethics Committee of the Clinica Fornaca di Sessant (usage of lipoaspiration product, protocol number EC20121008) and the CIR-Dental School of the University of Turin (harvest of deciduous teeth, protocol number CIR20121022). Written informed consent was always obtained from donors. As previously reported [17], ASCs were isolated and expanded from the lipoaspirates of ten different donors (mean age 25.6 ± 4.5 years, male/female = 5/5) at the Clinica Fornaca di Sessant, Turin. Following Miura's protocol [20], SHED were extracted from integer exfoliated deciduous teeth that were collected from ten children (9.2 ± 2.2 years) undergoing tooth extraction at the CIR-Dental School. For both cell types, the non-adherent cell populations were removed after 48 h and the adherent cell layer was washed twice with fresh medium. ASCs and SHED were maintained in growth medium (GM) consisting of RPMI (EuroClone S.p.A., Milan, Italy) enriched with sodium pyruvate and supplemented with 10% fetal bovine serum (Gibco BRL, Milan, Italy), penicillin-streptomycin 1:100, at 37 • C in humidified atmosphere with 5% CO 2 . Cells were then continuously cultured following their harvest until fourth passage. To induce osteogenic differentiation, cells were cultured in "osteogenic medium" (OM) by supplementing GM with 10 mM ß-glycerophosphate, 50 µg/mL ascorbic acid, and 0.02 mg/mL dexamethasone [52,53].

Cell Morphology
As described elsewhere [54][55][56], cells were seeded at a concentration of 5 × 10 3 cells/well in a 24-well plate for 1 day. After fixing in 4% paraformaldehyde in phosphate buffered saline (PBS), cells were stained with rhodamine phalloidin and DAPI (Life Technologies, Milan, Italy) to mark actin network and nuclei, respectively. Images were acquired with a Nikon Eclipse Ti-E microscope using different a Nikon Plan 20X/0,10 (Nikon Instruments, Amsterdam, The Netherlands).

Phenotype of ASCs and SHED
Cell surface markers of ASCs and SHED were analyzed by flow cytometry. In detail, cells were identified as CD105, CD44, CD73, and CD90 positive cells and negative for CD45 expression. Standard labelling protocol was performed with the following antibodies fluorochrome-conjugated and isotypic controls: human CD105 PE (Invitrogen, Camarillo, CA, USA), CD73 FITC (kindly provided by Prof. Malavasi, University of Turin), CD44 FITC, CD45 PerCP, IgG1 PE and IgG2a PerCP (Miltenyi Biotech, Bergisch Gladbach, Germany), and IgG1 FITC-conjugated (Immunostep). As further control, unstained cells were also examined. About 10 5 events/sample were used for capture with CellQuest software. Data were analyzed with FlowLogic software (Miltenyi Biotech).

Detection of Interleukins, Chemokines, and Growth Factors by Bio-Plex System
To quantify a panel of major signaling molecules released within their medium, ASCs and SHED were kept in GM for 3 days until confluence to assess the baseline. Cells were also kept either in GM or in osteodifferentiating medium for 14 days. The conditioned media thus produced by ASC and SHED after 2 hour starvation in RPMI were analyzed with the flexible Bio-Plex system (Bio-Rad Laboratories, Hercules, CA, USA), following the manufacturer's protocol, as described elsewhere [57,58]. At least two independent experiments were made per sample. The concentrations of the following specific biomolecules was measured: IL-6, IL-8, IL-10, IL-12, MCP-1, VEGF, GROα, HGF, MIF, β-NGF, SCF, and SDF-1α. Concentrations of the analytes are expressed in pg/mL.

Viability Assay
ASCs and SHED were plated at a density of 500 cells/well in 96-well culture dishes and the viability was assessed by CellTiter-Glo ® (Promega, Milan, Italy) according to the manufacturer's protocol at 1, 3, 7, and 14 days [59,60]. The CellTiter-Glo ® Luminescent Cell Viability Assay is a homogeneous method of determining the number of viable cells in culture based on quantitation of ATP as a marker of metabolically active cells. The amount of ATP is directly proportional to the number of viable cells in culture; for this reason, this assay can be used as an indicator of cell proliferation (as reported by manufacturer).

RNA Extraction and Real-time PCR Analysis
Total RNA was extracted using PureLink RNA Mini Kit (Ambion, Life Technologies Italy, Milan, Italy). For quantitative real-time polymerase chain reaction (qRT-PCR), 1 µg total RNA was transcribed into complementary DNA by MultiScribe ® Reverse Transcriptase (High-Capacity cDNA Reverse Transcription Kit, Thermo Fisher Scientific, Waltham, MA, USA) and PCR analysis was then assessed using TaqMan probes from Roche. Transcript abundance, normalized to 18s mRNA expression, is expressed as a fold increase over a calibrator sample. qRT-PCR was performed on a 7900HT Fast Real-Time PCR System (Applied Biosystems, Life Technologies Italy, Milan, Italy). Specific primers and probes were designed using the Universal Probe Library-Assay Design Center-Roche Life Science software (www.lifescience.roche.com) [61,62].

Alkaline Phosphatase Activity
Alkaline phosphatase activity (ALP) was determined colorimetrically as previously reported [63] and assessed at day 7. Cells were lysed with 0.05% Triton X-100 and incubated with the reagent solution containing phosphatase substrate (Sigma-Aldrich, Milan, Italy) at 37 • C for 15 min. A calibration curve of p-nitrophenol standards was always used. Alkaline phosphatase values were determined (Optical Density 405 nm) and normalized to the whole protein content, which was determined (Optical Density 562 nm) with a BCA™ Protein Assay (Thermo Fisher Scientific).

Osteocalcin Detection
Following the manufacturer's protocol, osteocalcin (OCN) was measured in GM and OM media by means of Osteocalcin Elisa kit (KAQ1381 Invitrogen Corporation, Waltham, MA, USA) at day 21.

Von Kossa Staining
ASCs and SHED were grown in six-plate wells until day 21 in OM. Cells were washed once with PBS, then they were fixed with 4% paraformaldehyde for 10 min at room temperature prior to being washed again with PBS. Calcium salts were stained after von Kossa (vK) as reported previously [64], while representative pictures were captured under light microscopy by an Olympus camera.

Statistical Analysis
Data were analyzed by the use of GraphPad Prism6 (GraphPad Software, Inc., La Jolla, CA, USA) and Microsoft Excel (Microsoft Corporation, Redmond, WA, USA). Each experiment was repeated at least three times. Statistical analysis was performed by using the Mann-Whitney non-parametric t-test. A p-value of <0.05 was considered significant [65,66].
qRT-PCR data were analyzed with GraphPad Prism 6 (GraphPad Software, Inc., La Jolla, CA, USA) by performing ordinary two-way ANOVA with Sidak's multiple comparisons test for grouped analyses or Mann Whitney test for column analyses.

Conclusions
In summary, these data suggest that undifferentiated ASCs and SHED may be preferable to committed ones for general cell therapy approaches, due to their higher paracrine activity. Osteoinduction significantly affects the cytokine, chemokine, and growth factor profile in a differential way, as SHED, differently from ASCs, developed a marked pro-angiogenic signature.

Conflicts of Interest:
The authors declare no conflict of interest. The founding sponsors had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the decision to publish the results.