Interfacing Sca-1pos Mesenchymal Stem Cells with Biocompatible Scaffolds with Different Chemical Composition and Geometry

An immortalized murine mesenchymal stem cell line (mTERT-MSC) enriched for Linneg/Sca-1pos fraction has been obtained through the transfection of MSC with murine TERT and single-cell isolation. Such cell line maintained the typical MSC self-renewal capacity and continuously expressed MSC phenotype. Moreover, mTERT-MSC retained the functional features of freshly isolated MSC in culture without evidence of senescence or spontaneous differentiation events. Thus, mTERT-MSC have been cultured onto PLA films, 30 and 100 μm PLA microbeads, and onto unpressed and pressed HYAFF-11 scaffolds. While the cells adhered preserving their morphology on PLA films, clusters of mTERT-MSC were detected on PLA beads and unpressed fibrous scaffolds. Finally, mTERT-MSC were not able to colonize the inner layers of pressed HYAFF-11. Nevertheless, such cell line displayed the ability to preserve Sca-1 expression and to retain multilineage potential when appropriately stimulated on all the scaffolds tested.


Introduction
Stem cells hold the potential to open up new and exciting perspectives to treat many degenerative diseases in the next future. Convincing evidences have been provided about the presence of multipotent progenitor cells within bone marrow and almost all organs [1], thus suggesting their implication in tissue homeostasis and their possible role as potential targets for innovative treatments. Therefore, the activation of resident stem cells in vivo by injecting growth factors or their administration into the damaged organs are currently investigated [2]. Alternatively, the engraftment of engineered tissue substitutes fabricated seeding stem cells on biodegradable, biocompatible scaffolds has been proposed as a more suitable solution to repair injured tissues [3,4]. In fact, fine-tuned scaffolds could provide physical support to stem cell adhesion, proliferation, and differentiation, while favoring the organization of newly-formed tissues [5].
Natural (fibrin, chitosan, collagen, hyaluronic acid, . . .) and synthetic (poly-lactic and poly-lactic-co-glycolic acid, poly-caprolacton, . . .) compounds have been proposed as components of scaffolds for tissue engineering. Natural compounds better resemble the properties of extracellular matrix and are usually derived from animals, thus being potential vehicles for infectious diseases. Conversely, synthetic polymers can be more easily processed and sterilized before the implant and their degradation controlled [6]. Natural and synthetic materials are used to fabricate scaffolds with different structure and geometry (i.e., solid scaffolds, microbeads, hydrogels, foams, fibers, etc.), suitable for different applications. For example, fibrous scaffolds made of polycaprolactone (PCL) have been proposed for cartilage reconstruction [7], hyaluronan-based materials Journal of Biomedicine and Biotechnology for vasculogenesis [8], and alginate hydrogels for cardiac repair [9]. Additionally, biodegradable microbeads are under investigation to deliver cells and/or growth factors to treat neurodegenerative disorders [6].
The optimal combination of accurately designed scaffolds and autologous stem cells could represent an extraordinary chance to treat several degenerative diseases. In particular, the ability of bone marrow-derived mesenchymal stem cells (MSC) to generate virtually all cell lineages [10] in vitro as well as to engraft and home in vivo [11,12] has raised great hope about their actual suitablity for clinical practice. Among the protocols so far adopted to purify MSC, the use of Stem Cell Antigen 1 (Sca-1) as a marker to enrich stem cells in a number of tissues is widely accepted [5,13,14]. In fact, Sca-1 expression has been associated to multipotency and self-renewal in bone marrow, skeletal muscles, dermis, heart, and liver [15]. However, Sca-1 pos mesenchymal stem cells isolated from murine bone marrow and cardiac stem cells from murine heart display multipotency in culture, but their differentiating potential on bidimensional and three-dimensional scaffolds made of synthetic polymers is not comparable. In fact, in our hands, bone marrow-derived MSC do not differentiate as a mere effect of scaffold physical factors, while cardiac stem cells in the same microenvironmental condition, express and assemble phenotype-specific proteins [5]. Therefore, the present study has been designed to challenge scaffolds with distinct chemical and structural composition and investigate their capability to affect Sca-1 pos MSC multilineage potential. On this purpose, polylactic acid (PLA) films and microbeads as well as two preparations of HYAFF-11, a hyaluronic acid derivative, have been utilized. However, in a preliminary step, since MSC undergo early senesce in vitro, an immortal line of murine Lin neg /Sca-1 pos MSC has been generated through telomerase transfection and used to challenge scaffolds composed by different polymers and displaying distinct structure. This demonstrated that Sca-1 antigen, although identified as a stemness marker, does not concur to regulate stem cell lifespan in culture, while confirming that the ectopic expression of telomerase reverse transcriptase (TERT), the catalytic subunit of telomerase, has the ability to extend MSC lifespan preserving stemness features [16][17][18].

Isolation and Culture of Murine Lin neg Sca-1 pos MSC.
Lineage negative, Sca-1 positive mesenchymal stem cells (Lin neg /Sca-1 pos MSC) were obtained from 6-week-old female C57/Bl mice femurs by 2 sets of magnetic cell sorting protocol (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany), as described elsewhere [5]. An aliquot of each separated cell subpopulation (Lin neg /Sca-1 pos and Lin neg /Sca-1 neg ) was then stained with anti-Biotin PE-conjugated secondary antibody and analyzed with fluorescence activated cell sorter. The Lin neg /Sca-1 pos fraction was resuspended in complete IMDM medium (Iscove's Modified Dulbecco medium, Cambrex Bio Science, Verviers, Belgium) sup-plemented with 10% FCS, 1× ITS, 100 IU/mL penicillin and 10 μg/mL streptomycin. This medium will be hereafter referred to as complete medium.

Construction of a Vector Expressing mTERT.
To produce a vector expressing mTERT, pcINeo plasmid (Promega, Madison, Wis, USA) was digested with EcoRI and the mTERT coding sequence, obtained from plasmid pGRN190 (a kind gift from Geron Corporation), was inserted into the same restriction site EcoRI. The resulting transfer plasmid was named pCINeo-mTERT.
Passage 10 (p10) Lin neg /Sca-1 pos MSC were plated at 50% confluence on day 0 and medium was changed on day 1, 4 hours before transfection. After transfection (12-16 hours), the calcium-phosphate containing medium was removed and cells were washed and incubated with fresh medium.

Single-Cell Clone Isolation.
To obtain single-cell clones, mTERT-MSC were plated into 96-well plate at a single-cell level by limiting-dilution method and cultured with conditioned medium. To obtain conditioned medium, mTERT-MSC supernatant was filtered through a 0.22 μm sterile filter and diluted 1 : 1 with fresh medium. Wells were inspected under light microscopy every day to exclude those containing more than 1 cell. Clones derived from a single cell were further cultured to 70-80% confluence and seeded into 1 well of a 24-well plate and thereafter serially reseeded in 6-well dishes, 25-cm 2 , and 75-cm 2 flasks.

Forward
Reverse 3CCCTAA (Biogen, Rome, Italy) sealed at the bottom of the tube by a wax barrier. Samples were incubated at 22 • C for 20 minutes to allow telomerase to extend TS primer, followed by a 31-cycle PCR amplification (Perkin Elmer Corp., Norwalk, Conn, USA) of the telomeric products. An internal control (IC) was included in the reaction mixture. Forty μL of PCR products were run on 10% nondenaturing Acrylamide gels. Gels were fixed in 0.5 M NaCl, 50% Ethanol, and 40 mM Sodium Acetate (pH 4.2) and then exposed to X-Ray Film (Kodak, Rochester, NY, USA). Band intensity was quantified by bidimensional densitometry, using a Bio-Rad scanning apparatus (Bio-Rad Imaging Densitometer, GS-670; Molecular Analyst Software, Richmond, Calif, USA).

Measurement of Population Doublings.
Cells were seeded at the concentration of 2 × 10 4 /cm 2 in correspondence of each passage. Cell viability was quantified by Trypan blue exclusion every second day. The assessment of proliferative capacity was performed by calculating the number of population doublings, using the formula log 10 (total no./start no.)/log 10 2.

MSC Differentiation
Protocols. mTERT-MSC, passage 1 (p1) and passage 10 (p10) MSC were plated onto 35 mm dishes (2 × 10 5 cells/dish) or onto the scaffolds and cultured in adipogenic or osteogenic medium (Cambrex Bio Science, Verviers, Belgium). The medium was changed every second day. Adipored (Cambrex Bio Science, Verviers, Belgium) or Alizarin Red S staining was used to assess the occurrence of adipogenic or osteoblastic differentiation after 10 days. Cells cultured in absence of induction mediums were used as negative controls. The presence of lipid vacuoles was visualized under fluorescence microscope, while the production of calcium deposits was confirmed in light microscopy.

RNA Extraction, Retrotranscription, and Semiquantitative
PCR. Total RNA was extracted by Trizol Reagent (GIBCO BRL, Gaithersburg, Md, USA). Retrotranscription was carried out with 2 μg of RNA for each sample using RT M-MLV (Invitrogen, Calif, USA) in the presence of random hexamers. Semiquantitative analysis of RNA expression was carried out by RT-PCR by comparing the control transcript (GAPDH) and the transcript of interest when their amplification was in the exponential phase. Total DNA was then extracted from the interphase and phenol phase according to manufacturer instruction. The primers used are reported in Table 1. PCR products were size-fractionated in 2% agarose gel electrophoresis.  nuclear extracts were prepared as previously described [17]. Cellular proteins from 5 × 10 6 viable cells were separated by 10% SDS-PAGE, transferred to Hybond-P membranes (Amersham Pharmacia Biotech, Piscataway, NJ, USA). Filters were probed as described above. Two mg of microbead preparation were centrifuged at 14000 rpm for 5 minutes and resuspended in complete medium. Afterwards, all the scaffolds were incubated for 4 hours with complete medium at 37 • C. Therefore, the medium was gently removed and mTERT-MSC seeded onto the scaffolds. The cells were seeded at the concentration of 5 × 10 5 /well on the microbeads, at 5 × 10 5 /cm 2 onto PLA films and HYAFF scaffolds in low attachment wells (Corning Life Sciences, Mass, USA).

Generation of Immortalized Sca-1 pos MSC Cell Line.
Lin neg /Sca-1 pos MSC were obtained by two subsequent sets of immunopurification of adherent bone marrow progenitors by anti-Lin-cocktail and-Sca-1+ antibodies (Miltenyi Biotec GmbH, Germany). Cells were cultured until the 10 th passage and then transfected with either pCINeo-mTERT expression or control vector pCINeo by Calcium Phosphate, as described in Section 2. Cells overexpressing mTERT were selected with G418, at the concentration of 300 μg/ml for 2 weeks. Single cell-derived colonies were isolated by limiting dilution and cultured in complete medium (Figure 1(a)). Telomerase ectopic transduction resulted in the accumulation of mTERT protein in transfected MSC (Figure 1(b)). As expected, mTERT-MSC showed increased levels of mTERT protein in the nuclear compartment, as compared to passage 1 MSC (Figure 1(b), right panel). Moreover, a progressive upregulation of functionally active telomerase (Figure 1(c)) was detected in mTERT-MSC (at p30 and p50), in TRAP assay, as compared to control cells at passages 1 (p1), 7 (p7), and 10 (p10). Cell proliferative capacity was assessed calculating the number of population doublings, using the formula log 10 (total no./start no.)/log 10 2. While control MSC underwent 30 population doublings before exiting proliferation and proceeding to senescence, mTERT-transduced cells showed a significant up-regulation in the number of population doublings (more than 300, Figure 1(d)), and are still in culture.

mTERT-MSC Retain Morphology, Phenotype, and Functional Features of Freshly Extracted Sca-1 pos MSC.
Proliferating undifferentiated MSC displayed typical fibroblastoid morphology and grew in a monolayer until confluence (Figure 2(a)), while they became senescent after 10-12 passages in culture, showing flattened morphology (Figure 2(b)). After the transfection, mTERT-MSC displayed the typical elongated morphology and the capacity to grow in a monolayer (Figure 2   line demonstrated that they expressed high levels of Sca-1, CD105, and low levels of MDR-1, CD90, FLK-1, and CD31 (Figure 2(f)). Finally, mTERT-MSC preserved multipotential capacity, as compared to freshly extracted (passage 1) MSC (Figure 2(g)). These features were retained for more than 150 passages in the same culture conditions (data not shown).  (Figure 3(a)). Therefore, the ability of mTERT to protect MSC from apoptosis following H 2 O 2 -induced oxidative stress was investigated. ( Figure 3(a)). Conversely, apoptosis in MSC overexpressing mTERT was significantly reduced after challenging cells with H 2 O 2 concentrations >5 μM, as compared to control cells. The oxidative DNA damage caused by reactive oxidative species (ROS) can be detected by evaluating the levels of 8-oxo-7,8-dihydro-2 -deoxyguanosines (8-oxodG) [18]. In this study, (FLARE Fragment Length Analysis using Repair Enzymes)-Comet assay, a modified version of the alkaline comet assay (single-cell gel electrophoresis), was used to detect and measure the H 2 O 2 -induced oxidative damage in control and mTERT-transduced MSC. DNA damage was measured as the mean Comet % tail of control MSC at passage 1 (p1 MSC) and p50 mTERT-MSC cells treated with increasing concentrations of H 2 O 2 . The overexpression of mTERT was able to protect MSC from DNA damage induced by H 2 O 2 at all concentrations tested (Figure 3(b)). Similar results were obtained using mTERT-MSC in correspondence of later passages (data not shown).

mTERT-MSC Behavior onto Injectable and Fibrous Scaffold of Biodegradable, Biocompatible Polymers. mTERT-MSC
line was used to test the possibility that different scaffolds could sustain stem cell adeshion, growth, and differentiation. On this purpose, p68 mTERT-MSC were seeded onto PLA films and microbeads (30 and 100 micron diameter) and onto pressed and unpressed HYAFF-11 felts. Cell adhesion was studied after 2, 4, and 6 days by hematoxylin-eosin staining. As shown in Figure 4(a), the cells were able to adhere to all the surfaces tested. Nevetheless, some significant differences were detectable: clusters of mTERT-MSC were found on PLA microbeads after 4 and 6 days. Moreover, the cells were not able to colonize the inner layers of pressed HYAFF-11 and formed a monolayer (but many aggregates wrapped single fibers at the end of the scaffold) which detached after 2 days. On the contrary, cells were found in the core of unpressed HYAFF-11 felts. Figure 4(b) shows that no significant differences in Sca-1 expression was detected after 6 days on all the matrices considered. This result was strengthened by the evidence that after 2 weeks on all the scaffolds tested, mTERT-MSC still retained multilineage capacity, as shown in Figure 5.

Discussion
The complex relationship between stem cells and polymeric scaffolds has been analyzed in few studies, so far [5,[19][20][21]. Scaffold chemistry, geometry, and manufacturing process play a pivotal role in directing Sca-1 pos cardiac resident stem cell (CSC) commitment towards the cardiac phenotype. However, scaffold properties exert different effects on distinct stem cell populations; in fact, bone marrow-derived mesenchymal stem cells (MSC) do not activate the same gene programs as CSC, when cultured in the same conditions (scaffold/culture medium) thus suggesting that MSC and CSC do not share all their functional features, the latter being more prone to acquire the cardiac phenotype when challenged by an adequate microenvironment [5].
The present study was designed to give an insight in Sca-1 pos MSC ability to retain their own multilineage potential when cultured on scaffolds with different chemical and physical properties. Unfortunately, cultured MSC proceed to senescence within few passages [22], indicating that the expression of Sca-1 antigen, which has been associated with self-renewal and multilineage potential in stem cell populations from virtually all the organs [15], is not sufficient to extend MSC lifespan in culture. For this reason, in a first step of the present investigation, a novel purified murine Sca-1 pos mesenchymal stem cell line (dubbed mTERT-MSC) has been generated by transfecting purified Sca-1 pos mesenchymal progenitors isolated from the murine bone marrow with the catalytic subunit of telomerase enzyme mtert. This procedure has already been proven effective in extending lifespan in embryonic [23] and adult stem cells [24][25][26]. Here, we confirm that the transfection of Sca-1 pos MSC with mTERT results in the generation of an MSC cell line (mTERT-MSC) permanently displaying undifferentiated morphology and phenotype, as well as multipotential ability, which are typical hallmarks of freshly isolated mesenchymal stem cells [10,27]. mTERT-transduced Sca-1 pos MSC showed upregulation of mTERT protein levels in the nuclear compartment, as compared to untransfected controls. This observation suggested that ectopic mTERT underwent the process of phosphorylation and activation by cell kinases [28], which allows its translocation to the nuclear compartment [29]. mTERT-MSC line behaved exactly like native, freshly isolated MSC and, thus, could be used as a suitable model to study mesenchymal stem cell culture onto a series of scaffolds, which have already proven effective in vitro [5,8,28] and are under evaluation for clinical applications. On this purpose, PLA films and microbeads, altogether with two different preparations of hyaluronan-based scaffolds have been challenged for their capacity to sustain mTERT-MSC adhesion, growth, and multilineage differentiation. The scaffolds used in the present study were chosen since they are representative of almost all classes of synthetic and natural matrices (injectable, solid, and fibrous) available  for tissue engineering. All the scaffolds used in the present study allowed mTERT-MSC adhesion without evidence of toxic effects, as demonstrated by the absence of floating cells within the well during the culture (data not shown).
Nevertheless, in vivo experiments demonstrated that PLA scaffold degradation can cause local pH alteration and limited immune response. On the contrary, hyaluronic acid degradation does not produce any adverse by-products, but can be vehicles for infectious diseases [8]. Interestingly, some clear differences in cell behavior on the matrices were noticed. As expected, cell distribution onto PLA films was uniform, with cells retaining their spindle-shaped morphology. On the contrary, cell adhesion onto PLA beads resulted in the formation of cell clusters, which made difficult any further quantitative analysis or longer culture. Notably, mTERT-MSC were able to migrate through the unpressed scaffolds and colonize the inner layers, whereas cells did not penetrate the surface of the pressed scaffolds. This phenomenon, already observed when using endothelial cells, has been explained suggesting that larger pores and an open structure favour cell migration within the scaffold [8]. In the present study, a significant difference was found in the diameter of fibers in pressed versus unpressed HYAFF-11 formulations, the former having nanoscale width, the latter showing fibers in the microscale range. The tight stacking of nanofiber network in pressed scaffolds could indeed represent an obstacle for cell colonization, as previously proposed [30]. Such evidence appears of great interest since the unpressed felts could be useful to obtain thick tissuesubstitutes, while the pressed ones appear more suitable to produce monolayers. Further investigations on the phenotype of mTERT-MSC clarified that no effects on mTERT-MSC line phenotype are induced by the scaffold per se, independently of biological cues, throughout the culture time analyzed. When cultured for up to 6 days on all the matrices considered, mTERT-MSC continuously expressed the stemness phenotype, as demonstrated by the maintenance of Sca-1 antigen expression at high rate. Moreover, mTERT-MSC demonstrated the ability to differentiate towards the adipogenic and osteogenic phenotypes onto all the scaffolds tested. This evidence is in contrast with the recent findings by Engler and collaborators, who demonstrated that matrix elasticity is sufficient to direct mesenchymal stem cell fate within few days [19]. Such a discrepancy could reflect a divergent behavior between fresh and immortalized mesenchymal stem cells, but could also be explained by still unknown scaffold features.