In COPD, PH is one of the most frequent complications associated with shorter survival rates [
4,
25]. PH has been recognized as one of the predictive factors that is connected with worse clinical outcomes [
1,
7,
27]. However, the pathophysiological origin of PH in COPD is mostly unknown. One of the main pathophysiological changes is the remodeling of pulmonary arteries [
27]. Previous investigations have shown that vascular alterations and PH precede lung emphysema development in human and animals [
20,
28,
29]. Several studies have indicated a possible role of S100A4, a member of the calcium binding proteins, in non-COPD associated forms of PH. Children with congenital heart disease show PH with pulmonary vascular lesions and an increase in S100A4 expression [
18]. Furthermore, it has been hypothesized that the S100A4 protein regulates the motility of cells by controlling the cytoskeletal dynamics and promoting SMC proliferation [
30]. In the current study we have shown an up-regulation of S100A4 mRNA in microdissected intrapulmonary arteries from explanted end-stage COPD patients. High S100A4 expression was observed in pulmonary arteries not only located in the media of occlusive arteries but also in neo-muscularized vessels with a diameter of ~50 μm. These findings may point to a role of S100A4 in vascular remodeling.
Patients with end-stage COPD and cor pulmonale have a frequent occurrence of hypoxemia in the lungs [
25]. Post-mortem studies and analysis of explant lungs have shown the accumulation of smooth muscle cells in the media layer, together with thickening and fibrosis of the intima in pulmonary muscular arteries in these patients [
4,
31]. Hypoxic conditions induce up-regulation of S100A4 that is correlated with a thickening of the media layer in intrapulmonary arteries [
19]. Up-regulation under hypoxia is induced via HIF transcription factors. While HIF-1β is constitutively expressed in many cell types, HIF-1α is rapidly degraded by ubiquitin-proteasome system under normoxia. Under hypoxic conditions, HIF-1α is stabilized and can form heterodimers with HIF-1β. Upon translocation to the nucleus, they bind to HREs in the promoter region of the target genes [
11,
32]. We and others have demonstrated that S100A4 transcription is enhanced via HIF stabilization and promoter binding [
33]. However, recent studies point out that remodeling is not exclusive to patients with advanced disease as it has also been shown in patients with mild COPD who do not exhibit hypoxaemia [
28,
29,
34]. Barberà et al. [
4] showed vascular remodeling in smokers with normal lung function or mild COPD-patients without hypoxia indicating that other mechanisms than hypoxaemia are causative for PH in COPD in earlier stages [
4].
Cigarette smoking is one of the most important risk factors for COPD [
1,
3]. There are few animal models which imitate the pathological changes seen in COPD [
35]. One of the most accepted models is the exposure of animals to cigarette smoke, which appears to be one of the best approximations to the human disease [
35]. Typical pathological changes of the human COPD such as emphysema, and PH are also seen in this model [
20,
34]. Similar to our observations in sections from COPD patients, mice exposed to 8 month cigarette smoke showed considerable up-regulation of S100A4 mRNA in intrapulmonary arteries, a time-point which largely reflects the characteristics of human COPD. Furthermore, strong immune reactivity for S100A4 protein in the vascular compartment, especially in the media layer was observed. As we have shown previously, in our model mice exposed to cigarette smoke do not suffer from hypoxemia [
20], the hypoxia stimulus for induction of S100A4 expression can be excluded in this model. This supports the hypothesis that S100A4 may be involved in early vascular remodeling even in non-hypoxic, mild COPD-stages. But what is the mechanism behind the up-regulation of S100A4 in absence of hypoxia? We could show that human S100A4 gene contains functional putative HREs in its promoters and that siRNA-knockdown of HIF-1/2 decreases S100A4 expression in human PASMC. It is well recognized that HIF-1/2 is not only stabilized under hypoxic conditions but also under normoxia in the presence of ROS [
26,
36‐
38]. Regarding the origin of ROS, Guo et al. showed that nicotine, a major component of cigarette smoke, induces HIF-1α expression via mitochondrial reactive oxygen species in human non-small cell lung cancer cells [
38]. In addition, ROS may be derived directly from mainstream smoke that exist mainly in the gaseous phase [
26,
38‐
40]. ROS may activate Erk-5 (BMK-1) via c-Src kinase, which is thought to be an activator kinase of HIF-1α [
26]. Alternatively, direct induction of HIF1α by nicotine via acetylcholine receptor-mediated signaling cascades, including the Ca
2+/calmodulin, c-Src, protein kinase C, phosphatidylinositol 3-kinase, MAP kinase/Erk 1/2 was shown [
41]. Finally, a HIF independent but ROS dependent regulation of S100A4 has already been postulated. ROS induces the translocation of phospho-ERK to the nucleus leading to GATA-4 phosphorylation with subsequent S100A4 production [
14]. Subsequent interaction with RAGE induces SMC proliferation and migration [
14,
42]. In line with this finding, we could observe increased mRNA RAGE expression in the pulmonary arteries of the mouse emphysema model. Although we could not detect differences in RAGE expression between COPD and donors, similar to others [
43,
44] we could show RAGE positivity in SMC layer. Of note, systemic soluble RAGE has been suggested as a biomarker as the severity of emphysema was associated with lower levels of sRAGE [
45]. Meloche J et al. demonstrated that S100A4 stimulation recapitulated the PAH phenotype of PASMC and that RAGE inhibition attenuated PH in vivo [
42]. However, whether the up-regulation of S100A4 in remodeled arteries of COPD lungs is signaled in absence of hypoxia via ROS and HIF has to be determined in further studies.