In the present study we show that distal lung fibroblasts respond to prostacyclin, and that prostacyclin may alter fibroblast activity and thereby remodeling processes. Interestingly, lung fibroblasts from patients with COPD had a higher synthesis of prostacyclin compared to control fibroblasts. The prostacyclin analog iloprost decreased collagen I synthesis and contractile capacity in both fibroblasts from control subjects and COPD, whereas alterations in proteoglycan production and proliferative rate were only present in fibroblasts from control subjects. The present data implicate an important issue that severe COPD patients may have a reduced repair mechanism in the ECM structure of the collagen network in the distal lung. Previous studies support our findings that changes in ECM synthesis are involved in pathologic conditions of COPD  and that fibroblasts from COPD patients may have a reduced or defective capacity of tissue repair [17, 18, 28]. Conversely, in this study, fibroblasts from COPD patients showed not only an altered production of prostacyclin and proteoglycans, but also a general decrease in proliferative rate and migratory capacity and an increased contractile phenotype compared to fibroblasts from control subjects. The capacity of fibroblasts to respond to an injury through the production or inhibition of mediators as TGF-β1 and prostacyclin may determine the nature of the repair responses involving ECM homeostasis. Notably, TGF-β1 is involved in remodeling processes in COPD through an activation of fibroblasts and induction of ECM production and may regulate proteoglycan synthesis [17, 29]. TGF-β1 expression has previously been shown to be increased in central airways  and in peripheral blood  from COPD patients. Also, fibroblasts from peribronchiolar areas of lung tissue from patients with severe emphysema have increased production of TGF-β1. However, in our study, we could not detect any significant differences in TGF-β1 synthesis between distal fibroblasts obtained from COPD patients and control subjects, indicating that the synthesis of TGF-β1 may be dependent on cellular origin and also location in the lung. In the present study, we could show that TGF-β1 increased the production of the ECM proteins collagen I and biglycan in distal lung fibroblasts derived from both controls and COPD patients. However, there was no difference in collagen I synthesis between fibroblasts from control and COPD patients. Corresponding with our data, Krimmer et al did not detect any differences in fibrillar collagen between non-COPD and COPD fibroblasts after TGF-β1 stimulation; neither did cigarette smoke extract affect the fibroblast ability to deposit collagen . Noordhoek et al did also not find any differences in collagen I synthesis between parenchymal fibroblasts from patients with mild emphysema and patients with severe emphysema , implicating that collagen I synthesis probably is preserved in fibroblasts from COPD patients. Interestingly, in the present study, prostacyclin synthesis was significantly increased after TGF-β1 stimulation in distal lung fibroblast from COPD patients. In line with these findings, TGF-β1 stimulation increased both COX expression and enhanced prostacyclin synthesis in a human lung fibroblast cell line . In contrast, decreased levels of prostacyclin production have been found in distal lung fibroblasts from patients with interstitial pulmonary fibrosis . These data imply that alterations in the levels of prostacyclin may be a marker of ongoing remodeling processes. Prostacyclin has previously been associated with tissue repair and remodeling processes by inhibiting profibrotic responses of fibroblasts [9, 10, 36] and attenuating pulmonary fibrosis in animal models [16, 37]. In the present study, the prostacyclin analog iloprost reduced collagen synthesis and subsequently attenuated the increased collagen production in response to TGF-β1. These results correspond to other findings where prostacyclin down regulated collagen synthesis in rat cardiac fibroblasts . Nonetheless, we could not detect any significant differences in collagen I production between fibroblasts from COPD patients or control subjects after iloprost treatment. Conversely, in the present study, iloprost balanced the changes in collagen I synthesis by altering the production of the collagen-associated proteoglycans decorin and biglycan in fibroblasts from control subjects, but iloprost had no effect on these proteoglycans in fibroblasts from COPD patients. Decorin is thought to be a negative regulator of TGF-β1 by binding and neutralizing significant amounts of this growth factor . Decorin and biglycan also shape and complement the collagen fibril structure, and decorin mediates the binding of collagen fibers . Our data indicate that fibroblasts from COPD patients may have a defective repair mechanism in the collagen network fibrillogenesis. Thus, high levels of prostacyclin could generate reduced collagen synthesis that is not regulated and stabilized by decorin or biglycan, which thereby may accelerate the formation of emphysematous tissue in COPD. In line with the findings of Hallgren et al , we found that biglycan synthesis was reduced by the distally-derived fibroblasts from severe COPD patients, whereas there were no differences in decorin synthesis before or after TGF-β1 stimulation between COPD and control fibroblasts. However, decorin synthesis has been shown to be decreased in distal fibroblasts from patients with severe emphysema  and decorin gene expression has also been shown to be decreased in centrally-derived lung fibroblasts from patients with severe COPD . Decreased decorin and biglycan expressions have also been shown in peribronchiolar areas in patients with severe pulmonary emphysema [17, 32]. One explanation to these findings could be that decorin is regulated differently at mRNA levels and protein levels with a higher turnover rate at protein levels . Mice that lack decorin have dysfunctional collagen fibrils with reduced tensile strength  and show altered lung mechanical properties, as enhanced lung compliance . Subsequently, decorin has been shown to reduce lung fibrosis induced by TGF-β1. Hypothetically, taken all these data together, fibroblasts from COPD patients may have an imbalance in the regulatory properties of the collagen network homeostasis, indicating that proteoglycan production is dysregulated in the collagen network assembly in response to iloprost or TGF-β1. Matrix metalloproteinases (MMPs) are essential for the degradation of ECM, and MMP-9 has been shown to be upregulated in severe COPD . Notably, MMP-9 release appears to be resistant to glucocorticoid therapy . Interestingly, prostacyclin treatment attenuated MMP-9 synthesis in mesangial cells  and in an animal model of cigarette smoke induced emphysema . Unfortunately, we could not detect any MMP-9 synthesis in the present study. Importantly, distally-derived fibroblasts from severe COPD patients demonstrated a more contractile phenotype in this study, probably due to enhanced ROCK1 activity , than fibroblasts from control subjects, and treatment with iloprost attenuated the contractile capacity to the same level as fibroblasts from control subjects in the present study. The inhibitory effect of prostacyclin analogs on fibroblast gel contractions has previously been shown in healthy lung fibroblasts and the response was mediated through cAMP activation of PKA . Proliferative rate was decreased by iloprost in control fibroblast, whereas COPD patients had a generally reduced proliferative rate that was not affected by iloprost treatment. Fibroblasts from COPD patients have also previously been reported to have reduced proliferative capacity [17, 49], which may contribute to the emphysema formation in the distal COPD lung. In the present study, control fibroblasts were obtained from healthy subjects of mixed age. Changes in ECM components in this study may reflect ongoing natural aging processes  and it has been shown that aging processes in the lung may occur independently of emphysema formation related to COPD pathogenesis . We could not detect any differences in studied parameters due to age within the control group; neither could we detect any differences in studied parameters due to different sampling techniques, transbronchial biopsies versus lung explants, implicating that the alterations presented in this study is probably linked to disease, and not to aging. Despite the limited numbers of observations in the present study, we could support the findings in a parallel study published by Hallgren et al  that the different sampling techniques and the age distribution in the two study populations did not interfere with the obtained results. Medical treatments may also influence the fibroblasts obtained from the COPD patients. It would have been an advantage if we had lung function data from all the controls. On the other hand, the donor lungs from the healthy individuals had been judged by the clinicians to be suitable for lung transplantation and we received the lungs due to the fact that they could not find any matching recipients at the moment. It is well known that smoking and ex-smoking may have an impact on fibroblast function. However, the fibroblasts obtained from the former smoker lung did not differ from the other fibroblast controls in the studied parameters, neither could Hallgren et al. find any differences . Studies on smokers and patients in different GOLD stages will be performed in the future to further investigate the importance of prostacyclin in remodeling processes.