Dental tissue-derived mesenchymal stem cells (MSCs) are a reliable cell source for dental tissue regeneration. However, the molecular mechanisms underlying the directed differentiation of MSCs remain unclear; thus, their use is limited. The histone demethylase, lysine (K)-specific demethylase 4B (KDM4B), plays critical roles in the osteogenic commitment of MSCs by up-regulating distal-less homeobox 2 (DLX2) expression. The DLX2 gene is highly expressed in dental tissue-derived MSCs but the roles of DLX2 in osteogenesis are unclear. Here, we investigate DLX2 function in stem cells from apical papilla (SCAPs). We found that, in vitro, DLX2 expression was up-regulated in SCAPs by adding BMP4 and by inducing osteogenesis. The knock-down of DLX2 in SCAPs decreased alkaline phosphatase (ALP) activity and mineralization. DLX2 depletion affected the mRNA expression of ALP, bone sialoprotein (BSP) and osteocalcin (OCN) and inhibited SCAP osteogenic differentiation in vitro. Over-expression of DLX2 enhanced ALP activity, mineralization and the expression of ALP, BSP and OCN in vitro. In addition, transplant experiments in nude mice confirmed that SCAP osteogenesis was triggered when DLX2 was activated. Furthermore, DLX2 expression led to the expression of the key transcription factor, osterix (OSX) but not to the expression of runt-related transcription factor 2 (RUNX2). Taken together, these results indicate that DLX2 is stimulated by BMP signaling and enhances SCAP osteogenic differentiation by up-regulating OSX. Thus, the activation of DLX2 signaling might improve tissue regeneration mediated by MSCs of dental origin. These results provide insight into the mechanism underlying the directed differentiation of MSCs of dental origin.

Class temporal speci¯cation is a kind of important program speci¯cations especially for objectoriented programs, which speci¯es that interface methods of a class should be called in a particular sequence. Currently, most existing approaches mine this kind of speci¯cations based on¯nite state automaton. Observed that¯nite state automaton is a kind of deterministic models with inability to tolerate noise. In this paper, we propose to mine class temporal speci¯cations relying on a probabilistic model extending from Markov chain. To the best of our knowledge, this is the¯rst work of learning speci¯cations from object-oriented programs dynamically based on probabilistic models. Di®erent from similar works, our technique does not require annotating programs. Additionally, it learns speci¯cations in an online mode, which can re¯ne existing models continuously. Above all, we talk about problems regarding noise and connectivity of mined models and a strategy of computing thresholds is proposed to resolve them. To investigate our technique's feasibility and e®ectiveness, we implemented our technique in a prototype tool ISpecMiner and used it to conduct several experiments. Results of the experiments show that our technique can deal with noise e®ectively and useful speci¯cations can be learned. Furthermore, our method of computing thresholds provides a strong assurance for mined models to be connected.

Determining how to select a subset of test cases with high-fault detection capability becomes a key issue in code-level regression testing. Cluster analysis has been proposed to deal with this issue. It partitions test cases into clusters based on the similarity of execution profiles. In previous studies, execution profiles were represented as binary or numeric vectors. The vector model only considers the number of times that a function or statement is executed. However, it ignores sequential, the relations and structural information between function calls. Therefore vector-based methods do not always generate satisfying results. In this study, the authors presented cluster analysis of three different types of structural profiles, that is, function execution sequence, function call sequence (FCS) and function call tree. They designed and conducted empirical studies on five medium-sized programs to validate the effects of different profiles on regression test case reduction. Experimental results illustrate that sequential, call relations and structural information can aid to further improve fault detection effectiveness. In view of cost-effectiveness, FCS is regarded as to be the optimal profile. Furthermore, cluster analysis of FCSs is comparable to the additional branch coverage reduction technique with respect to fault detection effectiveness.