Isolation and Characterization of Human Dental Pulp Derived Stem Cells by Using Media Containing Low Human Serum Percentage as Clinical Grade Substitutes for Bovine Serum

Adult stem cells have been proposed as an alternative to embryonic stem cells to study multilineage differentiation in vitro and to use in therapy. Current culture media for isolation and expansion of adult stem cells require the use of large amounts of animal sera, but animal-derived culture reagents give rise to some questions due to the real possibility of infections and severe immune reactions. For these reasons a clinical grade substitute to animal sera is needed. We tested the isolation, proliferation, morphology, stemness related marker expression, and osteoblastic differentiation potential of Dental Pulp Stem Cells (DPSC) in a chemically defined medium containing a low percentage of human serum, 1.25%, in comparison to a medium containing 10% Fetal Bovine Serum (FBS). DPSCs cultured in presence of our isolation/proliferation medium added with low HS percentage were obtained without immune-selection methods and showed high uniformity in the expression of stem cell markers, proliferated at higher rate, and demonstrated comparable osteoblastic potential with respect to DPSCs cultured in 10% FBS. In this study we demonstrated that a chemically defined medium added with low HS percentage, derived from autologous and heterologous sources, could be a valid substitute to FBS-containing media and should be helpful for adult stem cells clinical application.


Introduction
Transplantation of tissues and organs generated from allogeneic embryonic stem cells requires large manipulations and still carries many questions. Thus, although embryonic stem cell research provides a promising alternative solution to the problem of a limited supply of organs for transplantation, the problems and risks associated with the need for immunosuppression to sustain transplantation of allogeneic cells or tissue and questions on their safety, such as teratoma formation still remain [1]. Using cells from a post-natal individual, rather than an embryo, as a source of autologous or allogeneic stem cells would overcome the biological and clinical problems associated with the use of embryonic stem cells, as well as solve the ethical dilemma associated with embryonic stem cell research. A number of stem cells have been isolated from fully-developed organisms, particularly humans, but these cells culture protocols involve large use of animal sera [2], such as FBS, or horse serum and that is associated with many problems: the composition of animal serum is unknown and varies between batches, interfering with the reproducibility of experiments and they may be contaminated with viruses, mycoplasms, prions or other pathogenic, toxic or immunogenic agents [3][4][5][6]. Because of such safety risks, regulatory authorities discourage or prohibit the use of animal sera and other components for the production of biological products for human use [7]. For these reasons we developed and tested a chemically defined culture medium added with a small amount of autologous and heterologous human serum, which allowed us to isolate a highly proliferative population of dental pulp stem cells (DPSC), which expressed embryonic as well as mesenchymal stem cell markers and showed osteoblastic differentiation capacity comparable to a medium containing higher FBS amounts.

Isolation and culture of Dental Pulp Stem Cells
After written informed consent of donors' parents and ethics approval from the Ethics Committee of the Medical Faculty of Udine, dental pulps derived from normal exfoliated human deciduous teeth, of 5 to 9-year-old children (24 subjects), were extracted using a syringe needle and were transferred into 35-mm Petri dishes (Falcon, BD-Biosciences, San Jose, CA, USA). To test the best suitable HS percentage, capable of isolate and expand DPSCs, dental pulps were cultured in presence of an isolation/ proliferation medium, as described by Ferro et al. [8,9,10], supplemented with 2.5%-1.25%-0.5%-0.25% human serum (HS).
For comparison, dental pulps were also isolated and cultured in basal medium, composed of F-12 Coon's and Ambesi's modified (Gibco-Invitrogen, Carlsbad, CA), Medium-199 and CMRL-1066 (Sigma-Aldrich, St. Louis, MO, USA), added with growth factors alone or in basal medium supplemented with 1.25% human serum alone. Human serum was obtained after written informed consent of the donors. DPSCs were not subjected to any type of depletion techniques and when reached confluence were detached by trypsin (Sigma), and sub-cultured in 100 mm dishes at the density of 2610 3 cells/cm 2 . The culture was maintained semi-confluent in order to prevent the differentiation of the cells, and medium was changed every 3 days. 5610 4 DPSCs at passage 5 (P5), plated in triplicate, in 60 mm dishes, were used to generate growth curves in presence of media with or without different human serum percentages, as previously described, and were counted every day from day 1 to day 5, without medium changing.
DPSCs were also isolated and cultured in DMEM (Sigma) added with 10% FBS, 25 mg/ml gentamycin (Gibco) and in isolation/proliferation medium added with 1.25% heterologous human serum (C-HS), derived from commercially available human male AB plasma (Sigma).
To test and compare DPSCs proliferative capacity 5610 4 DPSCs at P5, isolated by using 1.25% HS, 10% FBS and 1.25% C-HS media, were plated in triplicate in 60 mm dishes and used for generate growth curves counting the cells at day 1, 3, 4, 5 with medium changing at day three. Human embryonic carcinoma stem cells (Ntera2), purchased from ATCC (ATCC-LGC, Milan, IT), were used as positive control for embryonic stem markers as suggested by Liedtke et al. [11], and were cultured according to Gallagher et al. [12]. Human osteoblast like cells, hOB, (ATCC-LGC, Milan, IT) were used as positive control for osteoblastic differentiation and were cultured by the method [13]. Human primary thyroid cells were cultured as already described by [14], and used as osteoblastic negative control. 1301 cell line, Tlymphoblastic leukemia, (Sigma) was maintained in RPMI 1640 medium (Gibco) supplemented with 10% FBS, 100 U/ml penicillin and 100 mg/ml streptomycin and was used as reference in flow-FISH analysis. Cells were counted in triplicate using a Neubauer chamber (Marienfeld GmbH, Lauda-Konigshofen, Ge) and 0.4% trypan blue (Sigma) solution was used to highlight nonviable cells. Population doubling times were calculated, in triplicate, during logarithmic growth phase by using doubling time software v1.0.10 (http://www.doubling-time.com) [15]. The number of total cell generations was obtained by dividing total culture time, expressed in hours and calculated from data obtained from P0 to P2, for the doubling time calculated from data obtained at P0 and P2, as previously reported. Cell doubling exponential rate and the number of total generations served us to evaluate the number of total cells starting from two progenitor cells. Then total cells, calculated from data obtained from P0 to P2, were divided by the number of total cells starting from two progenitor cells, estimating approximately the number of the primary culture progenitor cells. Media depletion time represents the time when the number of dead cells were approximately $10% with respect to the viable cells.

Telomeric Repeat Amplification Protocol (TRAP)-assay
Pellets obtained from 1610 6 of P5 DPSCs cultured in 1.25% HS, 10% FBS and 1.25% C-HS media, were washed once with PBS, re-pelleted and resuspended in 200 mL of 16 lysis Buffer. The detection of telomerase reverse transcriptase (TRT) activity was performed utilizing the TRAPeze kit (Chemicon) following manufacturer instruction; in addition we also used Ntera2 cell extracts as supplemental positive control.
The transcript amount of each gene was normalized to b-actin. Relative fold change in expression was calculated using the DDCT method (CT values,30) with respect to undifferentiated cells.

Identification of the best suitable HS percentage
In order to identify the lowest HS percentage capable, in conjunction with our medium, to permit the best suitable isolation and culture condition for DPSCs, we cultured dental pulps, extracted using a syringe needle, in presence of different HS percentages 2.5-1.25-0.5-0.25%. Obtained DPSCs were not subjected to any type of common selection techniques (immunodepletion, physical centrifugation or filtration and chemical depletion by erythrolysis); also we did not use murine feeder layers, fibronectin or other adhesion protein layers. From two to five weeks after plating, depending on HS percentages, DPSC were small, highly proliferative and exhibit a homogeneous, fibroblastoid morphology with scanty cytoplasm (Fig. 1A-D).
To evaluate media efficiency we used two indicators: 1population doubling time, 2-medium depletion time. Growth curves confirmed that DPSC proliferate at high rate both in media added with 1.25-2.5% HS with respect to 0.25-0.5% HS (level of significance p,0.05). We noted a gradual decreasing in plating efficiency with lowering serum and cell population doubling time of DPSCs growing in medium containing 1.25-2.5% HS was about 25.361.5 hours and 29.861.3 hours respectively ( Fig. 1E and Table 1) (p,0.05), while that of cells growing in medium containing in 0.5-0.25% HS and was 31.461.2, 31.462 hours respectively ( Fig. 1E and Table 1). Medium depletion time in tested conditions 2.5% HS, 1.25% HS, 0.5% HS and 0.25% HS was respectively 5days, 4days, 3days and 2days. On the contrary, DPSCs isolated and cultured in basal medium (F-12 Coon's and Ambesi's modified, Medium-199 and CMRL-1066) added with 1.25% HS alone evidenced a similar morphology (Fig. 1F), but population doubling time and medium depletion time were 14661 hours and 2days respectively ( Fig. 1E and Table 1). Instead we found that dental pulps cultured in isolation/ proliferation medium without HS did not develop colonies and did not proliferate. All together these data allowed us to identify the medium added with 1.25% HS as the best suitable alternative to media containing lower serum percentages, due to its capacity to permit higher proliferation rate, as well as its higher medium depletion time. For these reasons, and because our intentions were to save and reduce as much as possible HS percentage, we decided to use isolation/proliferation medium added with 1.25% HS in the following comparison experiments, even with respect to media containing higher HS percentage.

HS and FBS media comparison
Consequently, we compared our isolation/proliferation medium capabilities with a commercial medium added with 10% FBS.
In addition, we also tested and compared the properties of our medium substituting HS with 1.25% of a commercial human serum (C-HS) which can be more easily accessible, facilitating the establishment of a large scale up production process. Morphologically cells cultured in 1.25% HS ( Figure 1G) were more homogenous with fibroblastic shape with respect to cells cultured in 10% FBS ( Figure 1H) and in 1.25% C-HS (I). Growth curves confirmed that DPSC proliferate at a similar rate in our medium added with 1.25% HS, 2862 hours, as well as in medium added with 1.25% C-HS, 31.562 hours, and both proliferate at higher rate with respect to 10% FBS medium, 4562.5 hours (Fig. 1J and Table 2) (p,0.05). The estimated number of stem progenitor cells in all primary culture conditions varied from 80 to 800. Moreover we evidenced that in both culture conditions the presence of a low HS percentage permitted to DPSC to be detached more quickly with respect to the same cells cultured in 10% FBS, approximately 25% lesser, without compromising cell adhesion.

Telomere and TRT activity assays
Several reports have demonstrated the role of telomere length and telomerase activity in stem cell self-renewing, ageing and mobilization processes [21,22]. In order to verify whether cells growing under our experimental conditions possessed telomerase activity, a TRAP assay was performed observing that cells displayed in both culture conditions a low but present TRT activity 1560.7% in 1.25% HS, 1460.8% in 10% FBS, and 14.860.5% in 1.25% C-HS with respect to Ntera2 cells, and 1760.5% in 1.25% HS, 1560.3% in 10% FBS and 1760.8% in 1.25% C-HS with respect to manufacturer positive control (Fig. 4A and Table 4) (p,0.05); Telomere length of DPSCs, as assessed by Flow-FISH analysis, was, relatively to the telomeric length of the 1301 cell line, and was1861.1% (R7) (Fig. 4C) in 1.25% HS, 1760.9% (R7) (Fig. 4E) in 10% FBS and 17.161.1% (R7) (Fig. 4G) in 1.25% C-HS respectively. For comparison, the average

Osteogenic induction
In order to compare differentiation properties, proliferating DPSCs cultured in 1.25% HS, 10% FBS and 1.25% C-HS media were osteo-induced for three weeks. During the osteoblastic inductive period cells changed their fibroblastoid morphology, developing an asymmetric shape with an enlarged end (Fig. 5A, B,  C), as previously demonstrated by Ferro et al. [8] In addition, the osteo-specific genes expression pattern for alkaline phosphatase (Alp) (D), collagen type I (Coll-I) (E), osteocalcin (Osc) (F), osteonectin (Osn) (G), osteopontin (Osp) (H), and runt-related transcription factor 2 (RUNX2) transcript variant 2 (I) were analyzed at mRNA level both after one and three weeks of osteo-induction.
Real Time PCR data, expressed as differentiated over undifferentiated DPSC, showed an increased and comparable expression, both at one and three weeks of osteoblastic induction, for all the tested markers with no significant differences deriving from the different proliferating conditions, as evidenced in Table 5 and Figure 5D-H.

Discussion
Since early 20th century scientists have been searching for methods to allow the isolation and growth of tissues and cells outside of the body. In the late 1940's the first cell line (HeLa) was cultivated in vitro in a fluid mixture of chicken plasma, bovine  Serum provides all of the growth factors, vitamins, co-factors, hormones, attachment factors (fibronectin, laminin), transport factors (albumin, globulin, transferrin), nutrients (nucleosides, amino acids, fatty acids, lipids), trace elements and other factors which limit free radicals, toxins and heavy metals. Serum is a very complex product and only a small percentage of the components have been fully identified. For this reason, and in the absence of a valid alternative, it remains the most effective growth product for cell culture available today.
Most sera used in cell culture are from animal, mainly bovine origin [2]. This brings some disadvantages such as antibodies which may impair or damage cell growth, the possibility of presence of adventitious animal viruses and the possible contamination with endotoxins and mycoplasmas which can damage fragile cell lines.
With respect to the pathogenic risks due to addition of FBS in culture media, autologous HS is considered a safer alternative excluding the transfer of animal derived infections and related immunogenic reactions.
Therefore, to safely produce DPSCs for clinical applications, we formulated and tested an isolation/proliferation media, reducing as much as possible serum percentage presence and substituting it by adding specific cytokines and growth factors in order to obtain a well-defined composition.
It is likely that DPSC represent a small subpopulation of dental pulp resident cells that, under experimental conditions, were predominantly and selectively proliferating [8,16,25,26]; moreover it has been reported that chemotactic gradient between the dental pulp fragment and the culture medium served as a vector directing the cells toward what is perceived as a site of injury, leading to their continuous and selective migration to the Petri dish [16]. For these reasons our approach has been developed in order to simplify the isolation procedure and overall to test the effective  media capacity to isolate stem cells, starting from a source which contains a low number of stem progenitors as dental pulps. The high expression of markers found in embryonic stem cells, adult stem cells, the high proliferation rate, the TRT activity as well as the relatively long telomere presence, evidence that DPSCs have an expression profile that partially overlaps with either ES and adult stem cells and confirms their undifferentiated state [17,27].
In addition our data demonstrate that osteoblastic differentiation potential is not negatively affected by our culture conditions, as evidenced by the Real Time PCR data and as previously demonstrated by us using the same methods and culture medium [8]. Such high proliferation rate, phenotype and differentiation capacity are consistent with results obtained using high 10-20% FBS percentages [16,25,26], demonstrating that a population of adult stem cells derived from human dental pulps could be obtained using a chemically defined medium which contains low HS percentage. More specifically, data evidenced that the coordinated action of the growth factors and the low HS percentage are responsible for this growth rate, because DPSCs cultured in presence of 1.25% HS alone or growth factors alone proliferate at lower rate or did not proliferate. The likeness in population purity between Immature Dental Pulp Stem cells (IDPSC) [16] and the population of cells described in the present study leads us to believe that they are similar. The difference consists in the use of a medium with very low human serum, derived from autologous and heterologous sources, which makes our medium more suited for human clinical applications. In addition medium capabilities were also confirmed starting from different adult stem cell sources, specifically from adipose tissue [9] and bone marrow [10]. In conclusion this medium hold strong promise in clinical reparative medicine for the treatment of degenerative or inherited diseases and are free of the ethical concerns raised by the use of ES. Moreover these data confirm that, in vision of a robust scale up process, even a commercial human serum, which can be more easily accessible, could be used to obtain similar results. Autologous ex vivo expanded adult stem cells could be used for implantation aimed to repair damaged, aged or diseased tissues and organs. Finally the ability to stably transduce DPSC cells with specific genes would also enable the genetic manipulation of DPSC autologous cells for the treatment of degenerative and congenital disorders [28].