Background
The heart as the major component of the cardiovascular system (CVS), pumps blood into the pulmonary and systemic circulation. The right part of the heart receives deoxygenated blood from the peripheral tissues through the cava veins and pumps it into the pulmonary trunk (PT) by low-pressure pulmonary circulation. It is well known that the lumen of,
e.g. carotid and coronary arteries may be stenosed due to atherosclerosis and their normal bloodstream is affected. However, the mechanisms of development of atherosclerotic lesions in the system of pulmonary arteries, i.e. low blood pressure vessels and low partial pressure of oxygen are still unclear. In general, atherosclerosis is an inflammatory disease, characterized by the chronic accumulation of inflammatory cells, storage of lipids and fibrous component in the innermost layer of the arterial wall [
1]. Increased serum low-density lipoprotein (LDL) cholesterol in the form of modified/oxidized LDL (oxLDL) is believed to play a key role during all stages of the disease by regulating the expression of chemokines and proinflammatory cytokines in the arterial wall [
1]. Pulmonary artery atherosclerosis is a frequent autopsy finding, associated with different clinical conditions, and an important predictor of aortic atherosclerosis, ventricular hypertrophy and pulmonary embolization [
2,
3]. In this context, pulmonary embolism (PE) represents the occlusion of the pulmonary artery branches by a thrombus (blood clot) that has traveled from elsewhere through the bloodstream. PE is a common, lethal disease with considerably high morbidity and mortality [
4]. Atherosclerotic lesions in the arterial tree often occur especially at bifurcations, i.e. at sites exposed to turbulent blood flow leading to mechanical shear stress on the vessel wall that affects the endothelial cell homeostasis by e.g. pro-inflammatory activation [
3]. Interestingly, lesions do not develop in veins under normal environment of low pressure and high flow, but when veins are used as arterial bypass, e.g. aortocoronary venous bypass, they may develop atherosclerotic lesions, probably because they are subjected to high pressure [
5,
6]. The precise mechanisms leading to the development and progression of atherosclerotic lesions still remain largely unknown, but the regulation by growth factors may have an impact in lesion development. This is further supported by the observation that growth factor concentrations are frequently increased under pathological conditions and therefore giving information on the severity of the disease [
7]. However, there are limited data on the use of biomarkers in predicting morbidity in populations of patients e.g. with atherosclerosis or acute PE. In this context, growth differentiation factor-15 (GDF-15) is linked to e.g. cardiovascular diseases and is related to mortality in older adults [
8,
9]. Recent data indicate that in patients with acute PE, elevated concentrations of GDF-15 amongst others, are helpful to identify patients at risk of death during the acute phase of PE [
10]. Under normal conditions, GDF-15 is only weakly expressed in most tissues [
11], which are protected from inflammation and lesion development [
12]. GDF-15 is a distant and divergent member of the transforming growth factor (TGF)-β superfamily [
11,
13‐
15] with characteristics that most recently suggested GDF-15 as a potent biomarker for cardiac events and/or atherosclerotic diseases [
16,
17]. Previously, we have shown that GDF-15 is a factor implicated in several pathophysiological processes including autophagy, inflammation, chronic vascular diseases, cancer, ischemia, and atherosclerosis [
18‐
21]. However, GDF-15 is strongly upregulated under conditions associated with cellular stress such as tissue hypoxia/hyperoxia, inflammation, and oxidative stress [
22‐
25]. Moreover, GDF-15 significantly contributes to subclinical coronary heart disease (CHD) independently of established cardiovascular disease (CVD) risks [
26] and has been frequently associated with CVD [
27]. Interestingly, hypoxia contributes to the formation of many pathologies of the blood vessel wall, like atherosclerosis, aortic aneurysms, pulmonary artery stenosis, and chronic venous disease [
25,
28,
29]. In this context, elevated GDF-15 levels have been recently found during early chronic obstructive pulmonary disease (COPD) [
29‐
32]. Moreover, GDF-15 has a potentially protective role for endothelial cells by promoting the activation of HIF-1 (hypoxia-inducible factor-1), which has been found in human pulmonary microvascular endothelial cells and umbilical vein endothelial cells (HUVEC) under hypoxia [
33,
34]. Hypoxia of the vascular wall may be caused by inadequate oxygenation or increased cellular oxygen demand triggered by shear stress and increased hydrostatic pressure [
35]. Thus, induction of GDF-15 expression by hypoxia and shear stress in combination with its effects on cell proliferation and apoptosis suggests a functional role in pulmonary endothelial cells and thereby in the pathobiology of complex vascular lesions in pulmonary arterial hypertension (PAH) [
33,
36]. However, the cellular tissue sources, as well as detailed functional effects of GDF-15 in the CVS, have not been completely elucidated. Several human studies associate GDF-15 levels with CVD as, acute myocardial infarction [
37], hypertensive patients [
38] and hypertensive left ventricular hypertrophy [
39]. However, after prolonged hyperoxia and consequent lung injury, GDF-15 mRNA expression was also markedly induced and found up-regulated in the lungs of patients with PAH [
22,
33]. Moreover, plasma GDF-15 levels in hypertensive patients with left ventricular hypertrophy (LVH) were higher than those of hypertensive patients without LVH [
40]. Additionally, in hypertensive patients a positive correlation between plasma GDF-15 levels and LVH was found, suggesting that GDF-15 may be involved in the development of LVH hypertension [
40]. A recent study has also demonstrated elevated serum levels of GDF-15 in patients with idiopathic pulmonary arterial hypertension (IPAH) [
41].
In summary, GDF-15 seems to be involved in orchestrating atherosclerotic lesion progression in arteries [
20,
42]. Thus, this work aimed to characterize the influence of GDF-15 deficiency on the development, progression, and morphology of atherosclerotic plaques in blood vessels with low-oxygen blood and low blood pressure as the PT, in hypercholesterolemic ApoE-knockout mice. Additionally, it was examined whether atherosclerotic changes exist in a human post mortem sample of the pulmonary artery.
Discussion
Atherosclerosis is characterized by a multifactorial pathophysiology that affects different organs, particularly the heart, brain, peripheral artery system and, thus, is associated with the development of CVD. In humans, atherosclerotic lesions usually are found at the origins of tributaries, bifurcations, and curvatures of arteries [
49]. The nature of the disease could be explained by local disturbances in the blood flow leading to shear stress. The high and the low-shear areas have been considered as primary sites of atheroma formation in the arterial tree [
50‐
52]. However, it has been frequently observed that atherosclerotic lesions do not develop in the veins in their normal environment of low pressure and a high flow. Interestingly, it has been shown that the veins develop atherosclerotic lesions when they are used as arterial bypass grafts where they are subjected to high pressure [
5,
6]. Similarly, atherosclerotic lesions develop in the pulmonary arteries under pulmonary hypertension [
33,
53]. Related to this, in general, high blood pressure is a well-recognized risk factor in CVD, a phenomenon that fits well in the hypothesis of “arterial wall stress” where the stress is produced not by blood flow but by blood pressure. In a similar way, low level of oxygen in the blood may contribute to the pathogenesis of various diseases of the vascular wall [
28]. In this context, atherosclerosis in the pulmonary arteries, its branches, and after pulmonary hypertension are rarely affected by atherosclerosis and are not common in human [
54,
55]. Most recently it has been shown that pulmonary artery calcification was significantly greater in patients with suspicion of stable angina pectoris compared to healthy control by using quantitative 18F-sodium fluoride positron emission tomography/computed tomography (NaF-PET/CT), a method that has been most recently suggested to be of clinical use in the early detection of pulmonary artery atherosclerosis [
56]. In this regard, characterization of morphological differences between the lesions in pulmonary arteries and other vessels of the circulatory system may be of interest for early diagnosis and the use of different therapies. Recently, a sample of the pulmonary artery taken during the anatomy preparation course (Additional file
1: Fig. S1) has shown typical characteristics of advanced atherosclerotic lesions, with a cap-like coating of the plaque by SMC and fibrous cells, with SMC also diffusely distributed and extended foam cells within the plaque. Atherosclerotic processes analogously occurring in advanced human lesions are also seen in the BT of ApoE
−/− mice [
57]. We found that long-term feeding of ApoE
−/− mice with CED is accompanied by up regulation of GDF-15 in atherosclerotic lesions, whereas GDF-15 deficiency reduced lumen stenosis in the BT as well as 18FDG uptake in the aortic arch [
20]. However, not much is known about the influence of GDF-15 on the development of lesions in the pulmonary arteries or the PT. In this context, GDF-15 is involved in orchestrating atherosclerotic lesion progression by regulating apoptotic cell death and IL-6–dependent inflammatory responses to vascular injury [
20]. GDF-15 deficiency inhibits significantly the lumen stenosis in the BT and the aortic arch compared with ApoE
−/− mice [
20] as well as (by trend) in the PT, according to the present investigations. To characterize the morphological plaque composition in the PT, we performed (immuno)histochemical investigations. It is generally accepted that atherosclerotic lesions containing a high number of M2 MФ are more stable [
58,
59]. In mouse, at the early stages of atherosclerotic plaques M2 MФ are present whereas M1 MФ are the most frequent phenotype in the advanced lesions [
60,
61]. In this respect, our results show an increase of CD68
+ in the GDF-15
−/−ApoE
−/− compared to the ApoE
−/− mice in the PT supporting our previous work in which we have already demonstrated a similar effect in the BT [
20].
However, we found that the percentage of the Nc area is lower in GDF-15
−/−ApoE
−/− than in ApoE
−/− mice in BT and PT, a sign of instability in both kind of vessels. The presence of a lower density of MФ in the lesions of ApoE
−/− animals may be due to the fact that the Nc areas in this group are larger than in GDF-15
−/−ApoE
−/− mice. These similar observations in both kind of vessels (PT and BT) indicate independence of the blood oxygenation and/or the pressure and appear to be affected by GDF-15 (deficiency) after CED. In this context, innumerable evidence confirms that endothelial dysfunction is a characteristic of patients with hypertension [
62,
63]. Inflammation is a common mechanism related to endothelial dysfunction and there is a close relationship between oxidative stress and inflammation [
64]. Consequently, we have investigated COX-2 and IL-6 immunoreactivities in the PT, as well as in the BT [
20]. Unlike to our publication concerning BT [
20] we here found a low COX-2 and IL-6 immunoreactivities in the lesions of PT were predominantly localized in endothelial cells and the subendothelial space of PT in GDF-15
−/−ApoE
−/− and ApoE
−/− mice, but not within atherosclerotic plaque as we have found earlier in BT [
20]. It is well known that the endothelial lining maintains normal with no or low expression of proinflammatory factors under normal homeostatic conditions. The well-known cardiovascular risk factors including smoking, aging, hypercholesterolemia, hypertension, etc. are associated with alterations in endothelial function. These characteristic findings and the lower degree of lumen stenosis observed in PT in comparison with BT could be related to the lower blood pressure in the PT. Thus, it is tempting to speculate that these circumstances inhibit the growth of lesions, leading to an improved stability in PT than in BT after 20 weeks of CED.
Recent evidence shows that SMCs contribute to the formation of the majority of atheroma foam cells in ApoE
−/− mice fed with a Western diet or standard chow for longer periods [
65]. Sm-α-actin is an isoform typical of SMC, present in high amounts in vascular SMC and serves as a differentiation marker of SMC [
66]. In previous publication, we showed no difference in the percentage of sm-α-actin–positive cells in atherosclerotic lesions in the BT in both genotypes [
20], unlike what we found in the PT. In atherosclerotic plaques in the PT of GDF15
−/−ApoE
−/− mice we observed an increase of CD68
+ MΦ and sm-α-actin
+ cells. However, since foam cells may originate from both, monocytes/MΦ or SMC, a characterization and identification of the origin of the foam cells can provide more information on the effect of GDF-15 (deficiency) concerning the effect of the oxygenation level or the blood pressure on the plaque development. Survival and death of MФ are important factors that affect the lesion development and progression. In this regard, we have previously shown that oxLDL induces GDF-15 expression and apoptosis in human MФ [
44,
45]. Subsequently, we have shown for the first time that a consequence of GDF-15 deficiency results in inhibition of lumen stenosis in the BT of GDF15
−/−ApoE
−/− mice after 20 weeks CED [
20]. These effects were observed, regardless of inhibition of apoptosis as well as autophagy and an increase in cell density but without effect on proliferation [
20]. However, compared with BT, in the PT we observed a similar effect on autophagy but not on apoptosis. In atherosclerotic plaques in the PT of GDF15
−/−ApoE
−/− mice, it is likely that the increased apoptotic processes are responsible for the reduction of lumen stenosis, but without effecting cell density and, as in the BT, cell proliferation. This suggests that the PT has a different mechanism of lesion remodeling compared to the BT and the aortic arch, however, GDF-15 seem to be involved in this mechanism, too. In chronic inflammatory lesions, often under low oxygen concentrations, MΦ are abundant and adapted to this condition [
67]. According to this, we found more CD68
+ immunoreactivity in the PT of ApoE
−/− or GDF15
−/−ApoE
−/− mice than in BT [
20]. It has been described in human atherosclerotic lesions, that the hypoxic regions, e.g. the necrotic cores, are rich in foam cells and MΦ [
68]. In this context, it is well known that in humans and animal models the growth of atherosclerotic plaques is accompanied by hypoxia, which promotes atherosclerosis [
69,
70]. For many cell types, a major effect of hypoxia is the induction of apoptosis [
71]. Hypoxia is a known stimulus of inflammation, angiogenesis, and apoptosis for MΦ [
72]. MΦ begin to adopt a glycolytic metabolism allowing them to adapt readily when exposed to low oxygen conditions [
67]. In this regard, it is proposed that certain populations of monocyte/MΦ survive better under conditions of low oxygen, thereby contributing to their increased numbers at sites of chronic inflammation as tumors, myocardial infarcts, and atherosclerotic plaques [
67,
73]. This could be an explanation for an increased percentage of apoptotic cells and CD68
+ MФ in plaques of the PT than in the BT in both genotypes of mice. Despite the finding of an enhanced percentage of CD68
+ MФ in the PT compared to the BT, apoptosis levels are higher in the PT, suggesting that these apoptotic cells represent another type of cells, e.g. SMC or polarized MΦ. In vivo, in atherosclerotic lesions, the polarization of MΦ in M1 and M2 populations are increased during plaque progression, where also apoptotic cells are localized in rupture-prone areas and necrotic cores [
72]. According to this, we found in lesions of the PT an increase of apoptosis in GDF-15 deficient mice accompanied by a reduction of the lumen stenosis. In this respect, hypoxia can induce survival or death by apoptosis or necrosis, depending on the cellular and metabolic environment [
74,
75]. Related to this, GDF15 expression can increase in response to diverse extracellular stress signals, such as hypoxia/anoxia and inflammation [
7], as described in cardiomyocytes, where GDF15 protects against apoptosis and protects the heart against ischemia/reperfusion [
76]. The expression of GDF15 in cancer cell lines results in cell growth arrest and increased apoptosis, which suggests that GDF-15 may have antitumorigenic activity [
7]. This may resemble the physiological hypoxic environment within atherosclerotic lesions, however, we found an increase of apoptosis in plaques of GDF-15
−/−ApoE
−/− mice. Thus, it may be assumed that the balance or imbalance between proliferation, cell arrest, and apoptosis are critical factors in determining the plaque stadium and the development trend to a stable, unstable, or rupture-prone lesion. The specific response of the MΦ, polarization, apoptosis, necrosis, inflammation, perhaps depends on the ability of cells to adapt their metabolism to the plaque environment, e.g. hypoxia [
73,
74]. However, studies related to atherosclerotic regression models propose that MΦ apoptosis is the major pathway for its removal from the plaque, although, recent studies have suggested that MΦ can proliferate in the arterial wall and the plaque. [
77,
78].
Nevertheless, here we show for the first time a possible difference in the pattern of remodeling of the atherosclerotic lesion between the BT and PT in ApoE
−/−, as well as in GDF-15
−/−ApoE
−/− mice fed with CED and the importance of GDF-15 in this phenomenon. Inhibition of apoptosis may be anti-atherogenic and may thus be suggested as a therapeutic strategy to control plaque progression in the PT. Moreover, we have already shown that GDF-15 deficiency results in the activation of proapoptotic and at the same time, the induction of antiapoptotic genes in peritoneal MФ incubated with oxLDL in vitro [
20]. These data are consistent with our in vivo findings in the BT, showing a reduction in TUNEL- or APG5L/ATG-positive cells in atherosclerotic plaques of GDF-15
−/−ApoE
−/− mice [
20]. These data confirm the assumption that GDF-15 may be important in MФ death and the remodeling of atherosclerotic lesions as has been also postulated by others [
79‐
81] and that GDF-15 signaling may be a useful novel target for therapeutic intervention.
Atherosclerotic lesions in the pulmonary artery are frequently found in individuals with cardiopulmonary mechanisms of death, such as PE [
55]. An early prognosis of PE is a crucial clinical challenge, which would allow choosing the appropriate treatment and reducing the mortality rate. In this context, GDF-15 has been identified as a predictor of CVD and related also to acute PE [
82]. According to this, the serum level of GDF-15 was found to be significantly higher in patients with PE compared with controls [
83,
84]. In this regard, we unexpectedly found in the human pulmonary artery from our anatomy course, a few numbers of GDF-15
+ cells distributed, mainly in the tunica media (data not shown). This observation could be interpreted that the origin of circulating GDF-15 in patients with PE is probably not the pulmonary arteries.
GDF-15 expression can increase in response to diverse cellular stress signals, such as hypoxia/anoxia, inflammation, acute tissue injuries and tumoral processes [
7]. In this respect, increased expression GDF-15—as observed in lungs of smokers and patients with COPD contributes to cigarette smoke -induced pulmonary inflammation [
85]. GDF-15 increases during COPD exacerbation but the role in stable COPD is unknown [
26]. In this context, GDF-15 can be used as a systemic marker in patients with COPD, regardless of other cardiovascular risk factors [
26]. Therefore, GDF-15 may be a powerful new biomarker not only for cardiovascular but also for cardiopulmonary vascular disorders as well as a therapeutic target.