4 MATERIAL AND METHODS
6.1 Markers of inflammation and ECM remodelling in dogs with cardiac diseases
Activation of the inflammatory system in association with an imbalance of ECM remodelling has been reported in human patients with cardiac disease and CHF (Bradham et al., 2002a; Anker and von Haehling, 2004; Chen et al., 2008; Graham et al., 2008), whereas its role in canine cardiac diseases is not known. In the presented work, the expression of inflammatory and ECM markers was studied in dogs with cardiac diseases, both in the blood (Study I) and in the myocardium itself (Studies II and III). Dogs with cardiac diseases exhibited significantly higher circulating IL-1, IL-2, MMP-1, -3 and TIMP-3 mRNA levels than healthy control dogs, but significantly reduced TNF-α, TGF-β3, TIMP-1 and TIMP-2 levels, while no differences were seen for IL-6, IL-8, IL-10 and IFN-γ (Study I). In contrast, the expression of all markers of inflammation and ECM remodelling was significantly increased in the myocardium of dogs with cardiac and systemic diseases in comparison to controls, and a further increase of the inflammatory markers IL-1, IL-6,
IL-8, IL-10, TNF-α, IFN-γ and the ECM remodelling markers MMP-2, TIMP-1 and TIMP-2 was observed in dogs with cardiac diseases versus those with systemic diseases (Studies II and III). These results confirm those of previous studies that suggest that cardiac diseases are associated with an inflammatory state of the myocardium and cardiac remodelling (Anker and von Haehling, 2004; Chen et al., 2008). Cardiac remodelling was reflected by the gross post mortem results and clinically by ventricular dilatation or hypertrophy and myocardial dysfunction. The lack of inflammatory marker upregulation in the blood alongside the increase of inflammatory markers in myocardium suggests primary myocardial activation of the inflammatory system and not a myocardial response to circulating inflammatory mediators. Furthermore, the histological examination did not reveal substantial myocardial inflammatory cell infiltration (Studies II and III), which renders cardiomyocytes and cells of the ECM the likely source of the markers (Tyagi et al., 1993; Kuhl et al., 1996; Li et al., 1999; Frangogiannis, 2008; Souders et al., 2009).
The stimulation of cytokine, MMP and TIMP expression in cardiac diseases, which further stimulates their own and each other’s expression through auto- and paracrine loops, might be caused by haemodynamic abnormalities that result in increased cardiac stretch and activation of the neurohormonal system (Weber et al., 1994;
Opie, 2002; Heineke and Molkentin, 2006; Dobaczewski et al., 2010; Frangogiannis, 2012). IL-1, IL-2, MMP-1, -3, and TIMP-3 mRNA levels were elevated in both blood and myocardium of dogs with cardiac diseases, and also in the myocardium of dogs with systemic disease. This might indicate a generalised increased transcription of these markers in diseases with cardiac remodelling.
Dogs with DCM showed a further increase in myocardial MMP-2, TIMP-1 and TIMP-2 transcription, when compared with dogs suffering from other cardiac diseases. This was also reflected by the more severe cardiac histological changes in the heart. However, as these dogs were also in the most advanced disease stages (heart failure class D, in contrast to class B and C3 of dogs with other cardiac diseases), association of marker expression and of histological changes with the severity of the disease instead of the actual disease type appears likely.
Previous investigations found increased myocardial IL-6 transcription (Oyama and Chittur, 2006) as well as increased valvular TGF-β1, TGF-β3, MMP-1, MMP-3, TIMP-1, -2 and -3, reduced MMP-2 and variable MMP-9 production in dogs with
DVD (Oyama and Chittur, 2006; Aupperle et al., 2008; Aupperle et al., 2009b;
Ljungvall et al., 2011; Obayashi et al., 2011). The levels of circulating IL-2, IL-7 and IL-8 protein were reduced in dogs with DVD (Zois et al., 2012). In contrast to these results, studies I and III revealed elevated circulating IL-2 and myocardial MMP-2 mRNA levels in dogs with cardiac diseases. These differences in results are consistent with different material (valves versus myocardium) and diseases investigated, and the measurement of protein versus mRNA. The expression of TIMP-1 and TIMP-2 was increased in valvular tissue of dogs with DVD (Oyama and Chittur, 2006; Aupperle et al., 2009b) and in myocardial samples of dogs with DCM (Study III), which suggests that fibrosis might be an important component in both diseases.
Interestingly, not only pro-inflammatory markers, but also the anti-inflammatory cytokine IL-10 and the immune modulator IFN-γ were elevated in dogs with cardiac diseases (Study II). This suggests limitation of the inflammatory reaction by IL-10, since it inhibits the production of pro-inflammatory cytokines (Fiorentino et al., 1991;
Mosmann, 1994; Krishnamurthy et al., 2009). IL-10 is also known as a Treg cell product, and Treg cells are anti-inflammatory which reduces fibrogenesis (Kvakan et al., 2009; Tang et al., 2012). IFN-γ is associated with a Th1 cell polarisation, and Th1 cells are also suspected to limit fibrosis (Mosmann et al., 1986; Zhu et al., 2010; Wei, 2011). Accordingly, IL-10 and IFN-γ might have a cardioprotective function.
However, in the end-stage of cardiac diseases, the marked inflammatory reaction might override these regulatory components, which would indicate that the increase in IL-6, TNF-α and IL-10 observed in dogs in study II predicts a poor outcome, as reported in people with decompensated heart failure (Miettinen et al., 2008). On the other hand, the anti-fibrotic effects of IL-10 and IFN-γ might also reflect progressive cardiac remodelling.
In studies II and III, the expression levels in the different cardiac regions were compared. In dogs with cardiac and systemic diseases, atria generally showed higher transcription levels than ventricles. These differences matched with the observed extent of pathological changes identified by the histological examination in dogs with cardiac diseases (leukocyte infiltration, lipomatosis cordis, cardiac fibrosis, muscular hypertrophy of small arteries, cardiomyocyte necrosis) and confirm previous reports that suggested differences in the remodelling processes in atria and
ventricles (Hanna et al., 2004; Brundel et al., 2005). Furthermore, dogs with cardiac disease showed significantly higher atrial IL-1, IL-8, TNF-α, TGF-β1 and TGF-β3, and dogs with atrial fibrillation higher MMP-2, MMP-13 and lox transcription, than dogs with systemic diseases (Studies II and III). This increase in inflammatory and pro-fibrotic marker expression in the atria of dogs with cardiac diseases might contribute to atrial dilatation, atrial arrhythmia, cardiac dysfunction and therefore progression of disease (Khan et al., 2004; Khan and Sheppard, 2006; Mukherjee et al., 2006). In dogs with atrial fibrillation (which had been diagnosed in three dogs with DCM and one with DVD in the present study), MMP-2 and MMP-13 might cause ECM degradation which could allow the observed left atrial dilatation, and lox might cause increased collagen crosslinking (Mukherjee et al., 2006; Lopez et al., 2010), both factors known to contribute to the development of atrial fibrillation (Mukherjee et al., 2006; Adam et al., 2011).