Osteocytes up-regulate the terminal differentiation of pre-osteoblasts via gap junctions.

We examined cell-to-cell interaction between pre-osteoblasts and osteocytes using MC3T3-E1 and MLO-Y4, respectively. First, GFP expressing MC3T3-E1 (E1-GFP) cells were generated to isolate the cells from co-culture with MLO-Y4. No changes were observed in the expression of osteogenic transcription factors Runx2, Osterix, Dlx5 and Msx2, but expression of alkaline phosphatase (ALP) and bone sialoprotein (BSP) in E1-GFP co-cultured with MLO-Y4 was 300-400-fold greater than that in mono-cultured E1-GFP. In addition, mineralized nodule formation was drastically increased in co-cultured E1-GFP cells compared to mono-cultured cells. Patch clamp assay showed the presence of gap junctions between E1-GFP and MLO-Y4. Furthermore, when the gap junction inhibitor carbenoxolone (CBX) was added to the culture, increased expression of ALP and BSP in E1-GFP co-cultured with MLO-Y4 was suppressed. These results suggest that gap junction detected between pre-osteoblasts and osteocytes plays an important role on the terminal differentiation of pre-osteoblasts.

Establishment and validation of a rabbit model for in vivo pharmacodynamic screening of tachykinin NK2 antagonists.

We attempted to establish and validate an in vivo pharmacodynamic (PD) rabbit model to screen tachykinin NK(2) receptor (NK(2)-R) antagonists using pharmacological and pharmacokinetic (PK)/PD analyses. Under urethane anesthesia, changes in intracolonic pressure associated with intravenous (i.v.) administration of a selective NK(2)-R agonist, ßAla(8)-neurokinin A(4-10) (ßA-NKA), was monitored as a PD marker. The analgesic effects of NK(2)-R antagonists were evaluated by monitoring visceromotor response (VMR) to colorectal distension in a rabbit model of visceral hypersensitivity induced by intracolonic treatment of acetic acid. Intravenous administration of ßA-NKA induced transient colonic contractions dose-dependently, which were inhibited by the selective NK(2)-R antagonists in dose- and/or plasma concentration-dependent manners. The correlation between PD inhibition and plasma concentration normalized with the corresponding in vitro binding affinity was relatively high (r(2) = 0.61). Furthermore, the minimum effective doses on the VMR and ID(50) values calculated in the PD model were highly correlated (r(2) = 0.74). In conclusion, we newly established and validated a rabbit model of agonist-induced colonic contractions as a screening tool for NK(2)-R antagonists. In a drug discovery process, this PD model could enhance the therapeutic candidate selection for irritable bowel syndrome, pharmacologically connecting in vitro affinity for NK(2)-R with in vivo therapeutic efficacy.