Search

Vascular pericytes undergo osteogenic differentiation in vivo and in vitro and may, therefore, be involved in diseases involving ectopic calcification and osteogenesis. The purpose of this study was to identify factors that inhibit the entry of pericytes into this differentiation pathway. RNA was prepared from pericytes at confluence and after their osteogenic differentiation (mineralized nodules). Subtractive hybridization was conducted on polyA PCR-amplified RNA to isolate genes expressed by confluent pericytes that were downregulated in the mineralized nodules. The subtraction product was used to screen a pericyte cDNA library and one of the positive genes identified was Axl, the receptor tyrosine kinase. Northern and Western blotting confirmed that Axl was expressed by confluent cells and was downregulated in mineralized nodules. Western blot analysis demonstrated that confluent pericytes also secrete the Axl ligand, Gas6. Immunoprecipitation of confluent cell lysates with an anti-phosphotyrosine antibody followed by Western blotting using an anti-Axl antibody, demonstrated that Axl was active in confluent pericytes and that its activity could not be further enhanced by incubating the cells with recombinant Gas6. The addition of recombinant Axl-extracellular domain (ECD) to pericyte cultures inhibited the phosphorylation of Axl by endogenous Gas6 and enhanced the rate of nodule mineralization. These effects were inhibited by coincubation of pericytes with Axl-ECD and recombinant Gas6. Together these results demonstrate that activation of Axl inhibits the osteogenic differentiation of vascular pericytes.

We have identified immunologically the protein kinase C (PKC) isoforms present in rat mesenteric small arteries, defined their distribution between particulate and soluble fractions, and studied their involvement in phorbol ester-induced contraction. Our analysis revealed the presence of the CA(2+)-dependent PKCs (alpha and gamma), Ca(2+)-independent PKCs (delta and epsilon), and the atypical isoform (zeta). PKCbeta could not be detected, whereas PKCgamma is likely to be of neural origin. All isoforms exhibited different distributions. PKCalpha, PKCepsilon, and PKCzeta were found in both particulate and soluble fractions. In contrast, PKCdelta was mainly in the particulate fraction, and PKCgamma was in the soluble fraction. Phorbol esters, which activate PKC and cause smooth muscle contraction, downregulated only the alpha and delta isoforms. This was associated with a parallel loss of contractile response to phorbol ester. The force developed to submaximal concentrations of noradrenaline was decreased after phorbol dibutyrate pretreatment, although the sensitivity and maximal response were unchanged. Phorbol ester pretreatment did not affect the contractile response to vasopressin. The sensitivity to non-receptor-mediated contraction, caused by k+ in the presence of prazosin, was slightly reduced by 4 alpha- and 4 beta-phorbol ester pretreatment. Maximal tension in response to this agonist was not affected. We conclude that PKCalpha and/or PKCdelta is necessary for phorbol ester-mediated contraction but is not essential for noradrenaline-, vasopressin-, or k(+)-induced contraction, demonstrating differences in the mechanisms involved in the contractile response between these agents.