The roles of cyclic adenosine monophosphate- and cyclic guanosine monophosphate-dependent protein kinase pathways in hydrogen peroxide-induced contractility of microvascular lung pericytes.

BACKGROUND Sepsis and posttraumatic inflammatory processes are accompanied by definite changes in microvascular permeability, particularly in the lung. These permeability changes may occur because of damaged regulatory mechanisms at the level of the capillary wall. Pericytes are adventitial cells located within the basement membrane of capillaries. These cells contain multiple cytoplasmic processes that envelope endothelial cells, and are consequently thought to stabilize capillary walls and participate in microcirculation and endothelial cell permeability. Data from this laboratory and other laboratories have confirmed that pericytes are contractile cells, adding to the evidence that pericytes may influence or help regulate capillary permeability. We have already determined that hydrogen peroxide (H2O2) causes dose-dependent relaxation in microvascular lung pericytes (MLPs) at 10 minutes and, conversely, dose-dependent contraction at 30 minutes. It is the aim of this study to determine the mechanism of this biphasic contractile response. Specifically, we will determine whether cyclic adenosine monophosphate (cAMP)- or cyclic guanosine monophosphate (cGMP)-dependent protein kinase intracellular pathways are responsible for the hydrogen peroxide-induced contractility of MLPs. METHODS Rat MLPs were isolated by previously published protocol and cultured on collagen gel matrices. MLPs were pretreated with either ODQ, a soluble guanylate cyclase inhibitor (100 mumol/L), for 15 minutes; GKIP, a protein kinase G inhibitor (100 mumol/L), for 1 hour; SQ22536, an adenylate cyclase inhibitor (100 mumol/L), for 15 minutes; or H89, a protein kinase A inhibitor (10 mumol/L), for 1 hour. Hydrogen peroxide was then introduced to each MLP culture at 10 mumol/L, 100 mumol/L, and 1 mmol/L. After each of these treatments, the surface area of the collagen gels was digitally quantified at 10 and 30 minutes. RESULTS SQ22536 attenuated both relaxation at 10 minutes and the contraction seen at 30 minutes for all concentrations of H2O2. H89 caused a marked basal relaxation and prevented the cells from contracting at 30-minute exposures to all concentrations of H2O2. Both ODQ and GKIP attenuated the relaxation at 10 minutes but had no affect on the later contraction. CONCLUSION The cGMP-dependent protein kinase pathway is a mechanism for H2O2-induced relaxation of MLPs. Up-regulation of cAMP and cGMP is responsible for early H2O2-induced relaxation and late contraction. Protein kinase A (cAMP-dependent protein kinase pathway) may be an important intracellular signaling protein in the H2O2-induced contraction of MLPs or may be unable to down-regulate cAMP once inhibited. This evidence further supports the concept that there are separate intracellular pathways that regulate divergent cellular responses. This idea parallels the clinical concept of reversible and irreversible dysfunction of cellular processes in shock, and that the cellular dysfunction is initiated by separate intracellular pathways.