Interactions between cells are a crucial mechanism to correctly heal a wounded tissue. Myofibroblasts have a central role during healing but their means to communicate with other cells is unknown. Microparticles (MP) have demonstrated a potential role as mediators of cellular interactions during various diseases. We have analyzed the production of MP by normal (Wmyo) and pathological (hypertrophic scar, Hmyo) myofibroblasts and human dermal fibroblasts (Fb) when treated with serum or plasma as examples of body fluids. We have shown that the presence of these body fluids induced a very significant increase in MP production by Wmyo while no MP production was denoted for Hmyo and Fb. These effects were at least due to thermally sensitive protein(s) with a molecular mass >30 kDa. Furthermore, the increase in MP production was not linked to an increase in apoptotic Wmyo. MP characterization showed that VEGF and FGF2 were present in MP and that endothelial and (myo)fibroblast cell growth can be stimulated by MP treatment. We postulated that MP production by myofibroblasts could modulate mesenchymal cell growth and angiogenesis during normal healing. J. Cell. Physiol. 225: 734–740, 2010. © 2010 Wiley‐Liss, Inc.

During wound healing, angiogenesis plays a crucial role in inducing adequate perfusion of the new tissue, thereby allowing its survival. This angiogenic process contributes to the formation of granulation tissue, alongside myofibroblasts. Myofibroblasts are cells specialized in wound contraction and synthesis of new extracellular matrix. Fibroblasts, considered by some to be at the origin of myofibroblasts, have already been shown to promote neovascularization. Thus, we hypothesized that myofibroblasts play a key role during angiogenic development in wound healing. We isolated myofibroblasts from normal human skin wounds and dermal microvascular endothelial cells (HDMVEC) and fibroblasts from skin. Using an in vitro fibrin-based model, we compared the proangiogenic activity of wound myofibroblasts to that of fibroblasts in the presence of HDMVEC. By immunostaining with collagen IV antibodies, we observed the formation of a capillary network significantly more developed when HDMVEC were cultured with myofibroblasts compared to the network formed in the presence of fibroblasts. The differences between these cell types did not result from a differential secretion of Vascular Endothelial Growth Factor or basic Fibroblast Growth Factor. However, in the presence of myofibroblasts, a significant decrease in matrix metalloproteinase activity was observed. This finding was correlated with a significant increase in Tissue Inhibitor of MetalloProteinase (TIMP)-1 and TIMP-3. Furthermore, inhibition of TIMP-1 secretion using shRNA significantly decreased myofibroblasts induced angiogenesis. These results led to the hypothesis that normal wound myofibroblasts contribute to the vascular network development during wound healing. Our data emphasize the critical role of wound myofibroblasts during healing.

The development of tissue-engineered substitutes of substantial volume is closely associated with the need to ensure rapid vascularization upon grafting. Strategies promoting angiogenesis include the in vitro formation of capillary-like networks within engineered substitutes. We generated both connective and adipose tissues based on a cell sheet technology using human adipose-derived stromal cells. This study evaluates the morphology and extent of the capillary networks that developed upon seeding of human microvascular endothelial cells during tissue production. We posited that adipocyte presence/secretory products could modulate the resulting capillary network when compared to connective substitutes. Analyses including confocal imaging of CD31-labeled capillary-like networks indicated slight differences in their morphological appearance. However, the total volume occupied by the networks as well as the frequency distribution of the structure's volumes were similar between connective and adipose tissues. The average diameter of the capillary structures tended to be 20% higher in reconstructed adipose tissues. Quantification of pro-angiogenic molecules in conditioned media showed greater amounts of leptin (15×), angiopoietin-1 (3.4×) and HGF (1.7×) secreted from adipose than connective tissues at the time of endothelial cell seeding. However, this difference was attenuated during the following coculture period in endothelial cell-containing media, correlating with the minor differences noted between the networks. Taken together, we developed a protocol allowing reconstruction of both connective and adipose tissues featuring well-developed capillary networks in vitro. We performed a detailed characterization of the network architecture within engineered tissues that is relevant for graft assessment before implantation as well as for in vitro screening of angiogenic modulators using three-dimensional models.

No abstract

Testosterone can be converted into androstenedione (4-dione) by 17β-hydroxysteroid dehydrogenase (HSD) activity likely performed by 17β-HSD type 2. Our objective was to evaluate the rate of testosterone conversion to 4-dione as well as expression and localization of 17β-HSD type 2 in omental (OM) vs. subcutaneous (SC) adipose tissues of men. Formation of 4-dione from testosterone was significantly higher in homogenates (p ≤ 0.001) and explants (p ≤ 0.01) of OM than SC tissue. Microscopy analyses and biochemical assays in cell fractions localized the enzyme in the vasculature/endothelial cells of adipose tissues. Conversion of testosterone to 4-dione was weakly detected in most OM and/or SC preadipocyte cultures. Positive correlations were found between 17β-HSD type 2 activity in whole tissue and BMI or SC adipocyte diameter. We conclude that conversion of testosterone to 4-dione detected in abdominal adipose tissue is caused by 17β-HSD type 2 which is localized in the vasculature of the adipose compartment.