Stem cells from apical papilla (SCAP) are derived from the developing tissue at the apex of a tooth root termed apical papilla [1, 2]. These cells exhibit mesenchymal stem cell (MSC) properties with multi-lineage differentiation potential and they express several neural markers when grown in neurogenic cell culture, including βIII-tubulin, glutamic acid decarboxylase (GAD), NeuN, nestin, neurofilament medium chain (NFM), neuron-specific enolase (NSE), and 2', 3'-cyclic nucleotide-3'-phosphodiesterase (CNPase) [2, 3]. SCAP are considered a type of cell source for odontoblasts responsible for root development and they are capable of regenerating pulp and dentin tissues in vivo . SCAP are highly robust in terms of relatively high population doubling and telomerase activities [1, 2]. We have previously shown that three types of dental stem cells, SCAP, DPSCs (dental pulp stem cells) and SHED (stem cells from human exfoliated deciduous teeth) can be easily reprogrammed into induced pluripotent stem cells (iPSCs) at a higher reprogramming rate than dermal fibroblasts using Thomson's four factors, LIN28, NANOG, OCT4 and SOX2 and their vector system .
iPSCs have tremendous medical applications and are very similar to embryonic stem cells (ESCs) [6, 7]. However, it was realized that iPSCs generated by viral vector transduction either using Yamanaka's four factors, c-MYC, KLF4, OCT4, SOX2 [8, 9] or Thomson's four factors prevent iPSCs from being more similar to ESCs because the transgene-carrying iPSCs have a different profile at global gene expression and epigenetic levels, and have altered differentiation into functional cell types [10–12]. Furthermore, it is obvious that carrying oncogenes such as c-MYC raises a safety concern for their clinical use [13–16].
Tremendous efforts have been made to deliver the reprogramming factors without viral vector integration. The approaches include transient expression using adenoviral or nonviral vectors [17, 18], removing the integrated vectors using piggyBac transposition [19, 20], minicircle DNA , and non-integrating episomal vectors [15, 22, 23]. However, the reprogramming efficiencies both in human or mouse systems were very low ranging from 0.00005% to 0.039% and most cases were at the lower end. The approach of non-integrating episomal vectors reported by Yu et al.  required three individual plasmids carrying a total of seven factors, including the oncogene SV40, and has not been shown to reprogram cells successfully from adult donors. Using recombinant protein-based four factors [24, 25], synthesized mRNA , and Sendai virus  to generate iPSCs has also been reported. Unfortunately, the protein transduction method is extremely difficult, labor-intensive and time-consuming at present, and modifying Sendai virus vectors or preparing synthesized RNA is technically demanding.
Recently, a single lentiviral 'stem cell cassette' (STEMCCA) carrying all four Yamanaka's factors was developed for the efficient generation of iPSCs from mouse postnatal fibroblasts . Subsequently, the STEMCCA was modified into an excisable single polycistronic vector containing loxP sites. Most importantly, the excision of the vector system after reprogramming improved the iPSC differentiation potential . The STEMCCA-loxP was then humanized to carry four human reprogramming factors, OCT4, SOX2, KLF4, and c-MYC (designated as hSTEMCCA-loxP) and has been shown to reprogram human fibroblasts successfully . This excisable single polycistronic vector system provides a high efficiency in reprogramming fibroblasts into iPSCs (mouse: 0.5%; human: 0.1% to 1.5%). To utilize iPSCs for clinical applications or to understand the biology of these cells, removal of exogenously introduced genes in these cells is a critical step, given the premise that iPSCs are independent of exogenous reprogramming transgene expression and retain their pluripotency following factor withdrawal. Using the single vector STEMCCA and Cre/loxP system, transgene-free (TF) iPSCs can be generated which possess pluripotent characteristics [29, 30]. In the present study, we applied this system and technology to generate TF iPSCs from SCAP and examined their neurogenic potential as the first step toward generating and characterizing TF iPSCs from other oral cells/stem cells.