Dissecting the roles of endothelin, TGF-beta and GM-CSF on myofibroblast differentiation by keratinocytes
Myofibroblasts are specialized fibroblasts that contribute to wound healing by producing extracellular matrix and by contracting the granulation tissue. They appear in a phase of wound healing when the dermis strongly interacts with activated epidermal keratinocytes. Direct co-culture with keratinocytes upregulates TGFbeta activity and also induces fibroblast to differentiate into alpha-smooth muscle actin (alphaSMA)-positive myofibroblasts. TGF-beta activity alone cannot completely account for alphaSMA induction in these co-cultures, and here we analyze mechanical force generation, another potent inducer of myofibroblast differentiation in this model. Using deformable silicone substrates, we show that contractile activity of fibroblasts is already induced after 1-2-days of co-culture, when fibroblasts are generally alphaSMA negative. Endothelin-1 (ET-1), the most potent inducer of smooth muscle cell contraction, was up-regulated in co-cultures, while blocking ET-1 with the ET receptor inhibitor PD156252 inhibited contraction in these early co-cultures. In 4-5 days of co-culture, however, fibroblast contractile activity correlated with an increased expression of alphaSMA expression. Stimulation of fibroblast mono-cultures with ET-1 in a low serum medium did not induce alphaSMA expression; however, ET-1 did synergize with TGF-beta. Surprisingly, GM-CSF, another mediatorstimulating myofibroblast differentiation in granulation tissue, inhibited alphaSMA expression in fibroblasts, costimulated with TGF-beta and ET-1. GM-CSF activated NFkappaB, thus interfering with TGF-beta signaling. Blocking TGFbeta and ET-1 largely impaired alphaSMA induction in co-cultures at day 7 and, in combination, almost completely prevented alphaSMA induction. Our results dissect the roles of TGF-beta and ET-1 on mechanical force generation in keratinocyte-fibroblast co-cultures, and identify GM-CSF as an inducer of myofibroblasts acting indirectly.
2004
92
2
262
274
REVIEWED