Contribution of circulating biomarkers to unravel the role of extracellular matrix in hypertensive cardiac remodelling

Arterial hypertension is associated with alterations in the structure of the myocardium and the arterial wall (i.e., remodelling), resulting in disturbances of cardiac and vascular function that contribute to the development of hypertensive cardiomyopathy and hypertensive vascular disease, respectively. Central to the structural alterations are changes in the regulation of extracellular matrix (ECM), particularly changes in collagen turnover leading to exaggerated and disorganized deposition of thick and stiff fibres of collagen types I and III [1,2]. Whereas at the myocardial level, the diffuse excessive deposition of collagen tissue is involved in the deterioration of left ventricular function and the facilitation of ventricular arrhythmias [1], at the vascular wall level it contributes to arterial stiffening, and to increased peripheral resistance and organ ischemia [2]. Therefore, the study of cardiovascular collagen turnover has attracted the attention of the medical community during the past years. Synthesis of collagen fibres in the myocardium and the arterial wall occurs by fibroblasts, their activated phenotype known as myofibroblasts, and vascular smooth muscle cells differentiated to a secretory phenotype. In response to increased mechanical load, neurohumoral activation (renin–angiotensin–aldosterone system) and bioactive molecules (growth factors, cytokines, matricellular proteins), these cells augment the production of collagen precursors as well as of the enzymes responsible for the extracellular formation of collagen fibrils and fibres (i.e., procollagen proteinases and lysyl oxidase) (Fig. 1) [3]. Conversely, degradation of collagen fibres is primarily under the control of serine proteases and matrix metalloproteinases (MMPs), namely interstitial collagenase (MMP-1) and gelatinases A and B (MMP-2 and MMP-9, respectively), synthesized and secreted by fibroblasts, endothelial, and inflammatory cells [4] (Fig. 1). Although serine proteases (such as plasmin, trypsin, kallikrein, and cathepsin G) are capable of collagen degradation, their primary role in cardiac and vascular collagen turnover is thought to be activation of MMPs [5]. In addition to transcriptional and posttranslational regulation, MMP activity is regulated by endogenous inhibitors including α2-macroglobulin and the four distinct species of tissue inhibitors of MMPs (TIMPs 1–4) [4]. In fact, TIMPs bind noncovalently to the catalytic domain of MMPs in a 1 : 1 stoichiometry, preventing access to the substrate. In physiological conditions, collagen turnover is precisely regulated to maintain a finely balanced equilibrium between synthesis and degradation. Ultimately, the balance between collagen synthesis and degradation depends upon the control of MMP activity by TIMPs [6]. In pathologic conditions, such as hypertension, the excessive inhibition, as anticipated by a relative excess of TIMPs over MMPs, will reduce collagen degradation and facilitate the deposition of fibrotic tissue, whereas the insufficient inhibition, as anticipated by a relative abundance of MMPs over TIMPs, will increase collagen degradation and promote the disruption and loss of the physiological collagen scaffold [7,8].Fig. 1: No captions available.In this conceptual framework, some circulating molecules that meet several criteria to be considered as biomarkers of the above two processes have been proposed as potential useful tools to assess cardiovascular remodelling in arterial hypertension [9,10]. Particularly, biomarkers that can be used for collagen synthesis are the propeptides of procollagen types I (C-terminal and N-terminal propeptides) and III (N-terminal propeptide) (Fig. 1). For collagen degradation, collagen type I carboxy-terminal telopeptide, with MMPs (namely MMP-1, MMP-2, and MMP-9) and TIMPs (TIMP-1, TIMP-2, and TIMP-4) might be useful as biomarkers (Fig. 1). However, despite the available information on these biomarkers in hypertensive patients, a number of major issues remain that limit their clinical usefulness. In fact, based on current knowledge, it is not possible to determine prognostic, diagnostic, or predictive cut-off values for hypertensive cardiovascular damage. Although several studies show cut-off values, these values were not used in a prospective way. The cut-off points were determined retrospectively; with a receiver operating characteristic curve, the investigators try to obtain the optimum cut-off point with a significant prognostic value. However, to use the biomarkers in a prognostic or diagnostic way, more prospective research is definitely required. On the contrary, the correlation between circulating biomarker levels and both their expression at the myocardial and vascular levels and the amount of fibrotic tissue is another important part that has not been adequately addressed because histological investigation of myocardial and vascular wall fibrosis in patients with chronic cardiac diseases is difficult to address in clinical practice. Moreover, although noninvasive identification of diffuse myocardial fibrosis may be possible with evolving cardiac magnetic resonance and molecular techniques, they still require further investigation [11]. Another major limitation in the field of collagen biomarkers is the lack of systematic meta-analysis of the multiple clinical studies published in order to both analyse the discrepancies in the results among different studies and identify reproducible patterns of alterations in the biomarkers. Therefore, the time has come for collaborative research initiatives aimed to cover this aspect, as is the case for the study by Marchesi et al.[12] published in this issue of the Journal of Hypertension. The authors have performed a meta-analysis of articles available in MEDLINE and EMBASE databases up to July 2011, dealing with the measurement of circulating MMPs and TIMPs in hypertensive patients. The aim of the meta-analysis was to determine whether there are differences in the plasma levels of MMPs and TIMPs in hypertension compared with normotension, and whether plasma levels of MMPs and TIMPs are related to clinical parameters that become altered as a consequence of cardiovascular remodelling (i.e., left ventricular mass, aortic pulse wave velocity and augmentation index, and carotid artery intima–media thickness) in hypertension. After analysing 30 studies considered eligible, the authors found that plasma MMP-9 and TIMP-1 levels are higher in hypertensive patients without heart failure than in normotensive individuals. In addition, among hypertensive patients, plasma TIMP-1 levels are increased in patients with left ventricular hypertrophy (LVH) compared with patients without LVH. Finally, plasma MMP-2 levels were found increased in hypertensive patients with diastolic heart failure (DHF) compared with hypertensive patients without DHF. From their findings, Marchesi et al.[12] conclude that among the families of MMPs and TIMPs, TIMP-1, MMP-2, and MMP-9 have the potential to serve as circulating biomarkers of cardiovascular remodelling in hypertension. Furthermore, TIMP-1 may be considered as a potential biomarker of hypertensive cardiomyopathy. Despite a number of limitations (e.g., the small total number of individuals included in the analysed studies, the high heterogeneity among these studies, and the possibility of publication bias) that may challenge the robustness of the conclusions, the authors must be congratulated because their findings support the notion that the study of circulating biomarkers may unravel the role of ECM in the cardiac impact of arterial hypertension. Particularly, the authors provide a new piece of evidence on the linkage of TIMP-1 with hypertensive cardiomyopathy. In fact, a direct correlation has been reported between TIMP-1 concentration in coronary sinus blood and TIMP-1 concentration in antecubital vein blood in hypertensive patients with LVH after exclusion of other extracardiac sources, suggesting the potential cardiac origin of circulating TIMP-1 in these patients [13]. However, the association of the circulating levels of this molecule with its myocardial expression has not been yet demonstrated. Increased circulating TIMP-1 has been found to exhibit a good sensitivity and specificity for predicting diastolic dysfunction [14] and DHF [15] in hypertensive patients with LVH. Interestingly, it has been reported that circulating TIMP-1 is directly associated with left ventricular filling pressures in hypertensive patients with LVH and DHF [16], thus suggesting that an excess of TIMP-1 may contribute to the progressive deterioration of diastolic function in hypertensive patients with LVH. From a mechanistic point of view, the link of TIMP-1 with hypertensive cardiomyopathy may be approached from the perspective that it may contribute to myocardial remodelling via both fibrosis and cardiomyocyte hypertrophy. TIMP-1 binds to the active site of MMP-1, MMP-2, and MMP-9 in a 1 : 1 molar ratio and, thereby, an excess of TIMP-1 relative to these MMPs will prevent the access to their substrates for the subsequent proteolysis and degradation of collagen fibres [4]. In support of this possibility is the finding that an association exists between upregulation of TIMP-1 and decreased collagenolytic activity of MMP-1 in the hypertrophied and fibrotic left ventricle of different rat models of chronic pressure overload [17,18]. Furthermore, increased TIMP-1 expression in the setting of normal or reduced MMP-1 expression has been found in the fibrotic myocardium of patients with chronic pressure overload and asymptomatic diastolic dysfunction [19] or DHF [20], respectively. Furthermore, TIMP-1 may exert profibrotic actions that appear to be independent of any MMP inhibitory effect. For instance, TIMP-1 may stimulate growth and inhibit apoptosis of fibroblasts, as well as induce the phenotypic differentiation of these cells into myofibroblasts with high capacity to synthesize and secrete fibrillar collagen [21]. Finally, some findings suggest that TIMP-1 may act at the cardiomyocyte level, although the receptor-binding and the signalling pathway still remain unknown [22]. Interestingly, Meier et al.[23] recently demonstrated that norepinephrine-induced LVH in mice was accompanied by a time-dependent elevation of TIMP-1 in cardiomyocytes. This increase in TIMP-1 was shown to be necessary for the induction of the fetal gene program, a hallmark of cardiac hypertrophy [24]. In summary, the article by Marchesi et al.[12] provides important information supporting that TIMP-1 may be a biomarker of hypertensive cardiomyopathy. Nevertheless, additional studies are required to definitively cover in an integrated manner the following aspects: prospective validation of the incremental information provided by TIMP-1, combining it with standard biochemical markers (e.g. circulating natriuretic peptides) and emerging genetic and imaging markers in independent populations; assessment of the effects of their measurements on patient management and outcomes; and evaluation of the feasibility and cost-effectiveness of the application of this strategy in the community. It is important to stress that one added clinical value of a biomarker will be given by the possibility of targeting it therapeutically. In this regard, controlled underexpression of TIMP-1, while maintaining a balance in the TIMP-1–MMPs interactions, could be of interest in developing therapies for hypertensive cardiomyopathy in which TIMP-1 levels are increased [25]. It is challenging to explore whether this approach could offer a protective outcome in hypertensive cardiomyopathy. Acknowledgements This work was partially funded through the Unión Temporal de Empresas-Fundación para la Investigación Médica Aplicada (UTE-FIMA) agreement, the Instituto de Salud Carlos III, Ministry of Science and Innovation, Spain (cooperative project, Red Temática de Investigación Cooperativa en Enfermedades Cardiovasculares-RECAVA, grant RD06/0014/0008), and the European Union (MEDIA project grant HEALTH-F2-2010-261409 and EU-MASCARA project grant FP7-HEALTH-2011-278249). Conflicts of interest There are no conflicts of interest.