Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Myocardial Basis for Heart Failure: Role of Cardiac Interstitium
Acknowledgments
Dr. Francis Spinale is supported by the Research Service of the Department of Veterans Affairs. The author wishes to recognize the significant contributions of past MUSC and USC medical and graduate students that participated in our cardiovascular research program over the past two decades.
Extracellular matrix (ECM) remodeling in the context of heart failure (HF) is not readily described by a specific pathological stimulus, but rather the downstream consequence of multifactorial events. One useful categorical description that encompasses both the HF presentation and the key underlying physiological manifestations has emerged. Specifically, in patients with an HF presentation whereby quantitative imaging reveals a reduced left ventricular (LV) ejection fraction (EF <50%), this has been assigned the definition of HF with reduced EF (HFrEF) ( see Chapter 37 ), whereas the same clinical presentation accompanied by a relatively “preserved” EF (>50%) has been assigned the definition of HF with preserved EF (HFpEF) ( see Chapter 39 ). In general terms, the pathophysiological disease states that cause predominant and significant myocardial injury and loss of contractile units, such as ischemic heart disease and the dilated cardiomyopathies, would likely give rise to HFrEF. In contrast, HF that arises from an increase in long-standing LV afterload, such as hypertension or aortic valve disease, would be examples that give rise to HFpEF. It must be emphasized that in both categorical conditions, the underlying myocardial disease process is not the definition of HF, but only when they are sufficient to produce clinical symptoms do these definitions apply. In addition, these HF phenotypes are not mutually exclusive; thus patients may present with features of both impaired systolic function and diastolic function. Nevertheless, recent clinical trials have identified that a specific pharmacological intervention can be effective for HFrEF, but yield only equivocal results for HFpEF, underscoring the concept that the underlying structural basis for these HF phenotypes are distinctly different. Therefore, for the purposes of focus, the prototypical example of HFrEF as it applies to myocardial infarction (MI) and dilated cardiomyopathy (DCM) ( see also Chapter 20 ), and the prototypical example of HFpEF as it applies to hypertensive heart disease, will be utilized in this chapter.
One of the key features of HF irrespective of the category is that changes in LV, myocardial structure, and geometry occur and have been generically termed LV remodeling ( see also Chapter 12 ). Importantly, there are key features of the LV remodeling process that are present in either HFrEF or HFpEF and not only hold prognostic relevance but also provide potential clues to the underlying biology unique to each of these HF phenotypes. LV remodeling can be defined as changes in the geometry and function of the LV, which in turn is a summation of cellular and extracellular events. Moreover, it has become increasingly evident that the myocardial ECM is not a static structure, but rather a dynamic entity that may play a fundamental role in myocardial adaptation to pathological stress and thereby facilitate the remodeling process. A greater appreciation for the highly complex and dynamic nature of the ECM can be realized by myocardial imaging and direct interrogation of the interstitium, and will be discussed in upcoming sections. In both human and animal studies, it has been reported that alterations in the collagen interface, both in structure and composition, occur within the LV myocardium, which in turn may influence LV geometry. Moreover, dysregulation of key translational and posttranslational control points with respect to ECM have been directly linked to adverse LV remodeling and will be examined in this chapter. Therefore identification and understanding of the biological systems responsible for ECM synthesis and degradation within the myocardium hold particular relevance in the progression of both HFrEF and HFpEF. Accordingly, the purpose of this chapter is fourfold. First, we present a brief overview of myocardial ECM structure and biosynthesis. Second, we briefly demonstrate how the ECM is altered in specific disease states that give rise to HFrEF (MI, DCM) and HFpEF (pressure overload), and integrate studies identifying alterations at the genetic level. Third, we present a summation of how measuring dynamic changes in ECM remodeling through either plasma profiling or imaging can provide diagnostic/prognostic utility in HF. Fourth, we examine where the field of ECM biology may be moving in terms of developing therapeutics.
Myocardial Extracellular Matrix Structure and Composition
The myocardial ECM contains a fibrillar collagen network, a basement membrane, proteoglycans and glycosaminoglycans, and bioactive signaling molecules. The myocardial fibrillar collagens, such as collagen types I and III, ensure the structural integrity of adjoining myocytes, provide the means by which myocyte shortening is translated into overall LV pump function, and are essential for maintaining the alignment of myofibrils within the myocyte through a collagen-integrin-cytoskeletal-myofibril relation. While the fibrillar collagen matrix was initially considered to form a relatively static complex, it is now recognized that these structural proteins can undergo rapid degradation and fairly rapid turnover. Collagen fibril formation entails posttranslational modification. The carboxyterminal of the procollagen fibril is cleaved by a proteolytic reaction that results in a conformational change necessary for collagen fibril cross-linking and triple helix formation. A critical step in the proper formation and structural orientation of the fibrillar collagen matrix is collagen cross-linking. Interruption of collagen cross-linking has been clearly demonstrated to alter myocardial ECM structure and in turn LV geometry and function. Furthermore, alterations in fibrillar collagen cross-linking have been identified in myocardial samples taken from patients with end-stage HF. While the newly formed, uncross-linked collagen fibrils are vulnerable to degradation, the triple helical collagen fiber is resistant to nonspecific proteolysis, and further degradation requires specific enzymatic cleavage. During collagen cross-link formation, the carboxyterminal peptide is released into the vascular space. Collagen type I fiber formation results in the release of a 100-kDa procollagen type I carboxyterminal propeptide (PIP). Similarly, the formation of mature collagen III fibers results in the release of a 42 kDa procollagen peptide (PIIIP). As will be discussed, a number of associated proteins are necessary for the proper transport, assembly, and stability of the collagen fibril, and monitoring these collagen peptides and associated proteins has been demonstrated to provide diagnostic utility in terms of ECM remodeling and HF progression.
The integrins are a family of transmembrane proteins that serve multiple functions with respect to myocardial structure and function. The integrins form the binding interface with proteins comprising the basement membrane and therefore directly influencing myocyte growth and geometry. Moreover, the integrins coalesce at important structural sites within the myocyte, called costameres, which are composed of cytoskeletal proteins, such as alpha-actinin and vinculin, which form a key intracellular support network for contractile protein assembly and maintaining sarcomeric alignment. The myocyte costamere is also where the integrins appear to cluster and interdigitate with an intracellular signaling cascade system, such as focal adhesion kinase. Thus, disruptions of normal integrin–ECM interactions will likely result in significant changes in myocyte structure and function.
There are a number of extracellular proteins that comprise the basement membrane, such as collagen IV, fibronectin, and laminin. Thus, while fibrillar collagen has been the focus of most past studies, both basic and clinical, collagen is actually a modest component of the ECM. Thus, studies that evaluate both collagen content as well as other constituents of the ECM are beginning to emerge. For example, patients with DCM and mechanical assist devices demonstrate marked shifts in both fibrillar collagen and the basement membrane component, laminin. Representative images from this study are shown in Fig. 4.1 . It is the basement membrane that forms the anchoring points for the fibrillar matrix and contact points for other proteoglycans, such as chondroitin sulfate within the ECM. For example, the abundant binding of negatively charged unbranched glycosaminoglycans within chondroitin sulfate results in a molecule with a high osmotic activity. Accordingly, changes in the content and distribution of these proteoglycan will affect hydration within the extracellular space, and in turn directly influence myocardial compliance characteristics. Moreover, these highly charged molecules within the ECM result in the formation of a hydrated gel that serves as a reservoir for signaling molecules and bioactive peptides. Therefore, the ECM is an important determinant of extracellular receptor–ligand interactions. Another important function of the myocardial ECM is that it serves as a reservoir for critical biological signaling molecules that regulate myocardial structure and function. For example, tumor necrosis factor (TNF) is initially a membrane-bound molecule that requires proteolytic processing and release into the interstitial space in order to form ligand complexes with cognate receptors. Another example is transforming growth factor (TGF)-β, which exists in a latent form bound to the ECM, and requires proteolytic processing within the myocardial interstitium to become a competent signaling molecule. TGF signaling produces multiple cellular responses—most importantly, the stimulation of ECM protein synthesis. Thus, while this chapter will primarily focus upon the collagen fibrillar network, it is becoming recognized that the myocardial interstitium is a complex environment, which contains structural and signaling molecules that directly affect overall myocardial form and function.
The Myocardial Fibroblast
The most abundant cell type within the LV myocardium is the fibroblast. This is the predominant cell type that is responsible for maintaining ECM homeostasis in terms of collagen synthesis and degradation. This does not imply that the fibroblast is operative in isolation, as significant biomechanical and bioactive signals between the cardiocyte-fibroblast and endothelial cell–fibroblast interactions likely dictate the rate and type of ECM synthesis and degradation. What appears to occur in disease states, whereby enhanced fibrosis and/or ECM remodeling occurs, is a significant change in the fibroblast phenotype. This process, whereby a significant population of phenotypically differentiated fibroblasts arises with ECM remodeling, can been termed as a transdifferentiation process, and the resultant cell type has been defined as the myofibroblast. However, this does not imply that the normal fibroblast phenotype is extinguished, but rather there is likely increased proliferation of fibroblasts that do not express the myofibroblast phenotype. Thus, in ECM remodeling there is likely expansion and proliferation of both fibroblasts and myofibroblasts, which in turn will significantly affect the balance between collagen synthesis and degradation. In terms of myocardial ECM remodeling in HFrEF, changes in the synthesis-degradation axis occur, resulting in ECM instability ( Fig. 4.2 ). In contrast, with HFpEF, robust expansion of myofibroblasts occurs, resulting in increased ECM synthesis and accumulation ( see Fig. 4.2 ). The specific cell type of origin with respect to the myofibroblast remains unclear and may arise from a stem-cell type, pericyte, or a clonal expansion of endogenous fibroblasts. Whatever the case, the myofibroblast is not found in normal LV myocardium but is readily identifiable in the context of LV remodeling. The myofibroblast is typically identified through cellular constituents and by function. Specifically, the mature myofibroblast expresses alpha smooth muscle actin (SMA), whereas normal fibroblasts do not. In addition, cultured myofibroblasts demonstrate much higher proliferation, margination, invasion, and ECM contractile behavior.
One of the more active areas of research in the cancer field is that of mesenchymal cell transdifferentiation (MET), whereby mesenchymal cells under control of the local environment will transdifferentiate into a proliferative cancer type. There are significant similarities in the signaling pathways that are evoked during this mesenchymal transdifferentiation process to that of fibroblast–myofibroblast transdifferentiation. Since the myocardial fibroblast is a cell type of mesenchymal origin, it follows that the expression of specific transcription factors and cell markers would be operative similar to that of MET. There is compelling evidence to suggest that this fibroblast transdifferentiation is accompanied by an intermediate step or cell type, the proto-myofibroblast. This proto-myofibroblast will express unique cell markers, such as a splice variant of fibronectin—the extra-domain A (fibronectin ED-A). What is clear is that the canonical thought that the fibroblast is a single phenotypic entity must be revisited, and changes in the type, proliferation, and function of sub populations of fibroblasts likely contribute to specific forms of ECM remodeling, and in turn will influence myocardial form and function with HF.
Myocardial Extracellular Matrix Proteolytic Degradation: The Matrix Metalloproteinases
The matrix metalloproteinase (MMP) are a diverse family of zinc-dependent proteases that play a role in normal ECM turnover as well as in pathological tissue remodeling processes. Currently, over 25 distinct human MMPs have been identified and characterized. While initially thought to only cause ECM proteolysis, a highly diverse set of biological functions have been identified, which include cytokine processing and activation of pro-fibrotic signaling molecules. Moreover, MMP activity is highly dynamic within the human myocardial ECM. Specifically, using microdialysis techniques coupled with MMP fluorescent substrates, it has been identified that continuous MMP activity exists within the human myocardium under ambient conditions, which can increase significantly in a matter of seconds. The MMPs were historically classified into sub groups based upon substrate specificity and/or structure, and an informal nomenclature for some of the individual MMPs arose from these initial substrate studies. However, there is significant overlap in MMP proteolytic substrates, and a more rigid numerical classification is now utilized. Important MMPs in the context of myocardial ECM remodeling include the collagenases, such as MMP-1 and MMP-13; the stromelysins/matrilysins, which include MMP-3/MMP-7; the gelatinases, which include MMP-9 and MMP-2; and the membrane-type MMPs (MT-MMPs). Taken together, once the MMPs are activated, these enzymes can degrade all ECM components; therefore it is important that the activity of these enzymes is kept under tight control and interaction. The regulation of MMPs can take place at the level of transcription, posttranscription, posttranslation, and endogenous inhibition. These levels of control are briefly outlined in the next few paragraphs with examples of how this regulation is operative in the context of myocardial ECM remodeling.
Transcriptional Regulation of Extracellular Matrix Degradation; Matrix Metalloproteinase Polymorphisms
Transcriptional regulation of MMPs is primarily determined by upstream gene promoter activity whereby a number of intracellular signaling factors bind to specific sequences within the MMP promoter sequence. As such, there has been considerable interest in nucleic acid substitutions (i.e., polymorphisms) that occur within the MMP promoter regions and in relation to overall MMP levels, and most importantly in relation to cardiovascular outcomes. There have been a number of MMP polymorphisms identified in key MMP types, which include the collagenases (MMP-1, MMP-8), the gelatinases (MMP-2, MMP-9), and stromelysins (MMP-3). A brief synopsis of MMP polymorphisms with respect to cardiovascular remodeling processes and selected citations are provided in Table 4.1 . This summary table is by no means exhaustive but underscores the fact that several polymorphisms, primarily within the MMP promoter regions, have been identified and associated with subsets of patients at risk for cardiovascular events.
TABLE 4.1
Polymorphisms in Matrix Metalloproteinase Promoters
| Polymorphism | Biologic Effect | Clinical Association | References |
|---|---|---|---|
| MMP-9 | |||
| -1562 C/T | Promoter activity | Post-MI remodeling | 1, 2, 19–22, 87, 90 |
| Remodeling in DCM | 3 | ||
| -279 R/Q | Catalytic activity | Remodeling in DCM | 1, 3 |
| Risk of post-MI remodeling | 86, 91 | ||
| Differing pharmacologic response in CVD | 88 | ||
| Promoter Activity | CAD | 86, 101 | |
| TAA/AAA | 100, 103 | ||
| MMP-2 | |||
| -790 T/G | Transcription factor binding | CAD, HF, CVD, HTN, TAA | 23, 24, 92–94 |
| MMP-3 | |||
| 1171-5A/6A | Promoter activity | Post-MI remodeling | 4-7, 11, 22, 87, 96 |
| 1612-5A/6A | MVP | 97 | |
| Remodeling in DCM | 4, 89 | ||
| Promoter | CAD | 100, 104 | |
| HTN | 104 | ||
| AAA | 105, 106 | ||
| MMP-1 | |||
| -1607 1G/2G | Transcription factor binding | Risk of post-MI remodeling, HF, CVA | 9, 10, 89, 108 |
| -519-340 A/G-C/T | Promoter activity | Risk of post-MI remodeling, BAV | 12, 98, 99 |
| MMP-8 | |||
| Promoter | CVD | 107 | |
| TAA | 107 | ||
| MMP-12 | |||
| Promoter | CVD | 88 | |
AAA , Abdominal aortic aneurysm; CAD , coronary artery disease; CVA , cerebrovascular accident; CVD , cardiovascular disease; DCM , dilated cardiomyopathy; HTN , hypertension; MI , myocardial infarction; MVP , mitral valve prolapse; TAA , thoracic aortic aneurysm.
Several of the MMP types identified in Table 4.1 are associated with acute/chronic inflammation in that these proteases are expressed by inflammatory cells as well as being induced by mediators of inflammation, such as through cytokine signaling. For example, MMP-8 and MMP-9 are highly expressed by neutrophils and macrophages, and rapid induction of these MMP types through inflammatory signaling pathways has been well established. Coronary artery disease, and more specifically vulnerable coronary plaques, has been associated with inflammation and local activation of MMP-9 and MMP-8. With respect to MMP-9, a variant has been identified, which is a single base substitution between cytosine (C) and thymidine (T). The T allele results in a higher relative level of promoter activity when compared to the C allele. The presence of the MMP-9 T allele resulted in increased plasma levels of MMP-9. A naturally occurring variant in the MMP-3 promoter region are the 5A and 6A alleles; the 5A allele has been associated with increased MMP-3 promoter activity, and in turn, increased relative MMP-3 protein levels. In a study performed in over 2800 Japanese patients, the 5A polymorphism was associated with increased risk of MI, particularly in women. Mizon-Gerad and colleagues reported that homozygosity for the 5A allele in patients with nonischemic cardiomyopathy was associated with poor survival. In a preliminary clinical study, the addition of a guanine within the MMP-1 promoter region (-1607 1G/2G) was associated within an acceleration of adverse LV myocardial remodeling in patients post-MI. With respect to MMP-8, increased local MMP-8 levels and polymorphisms within the MMP-8 promoter region have been identified with coronary plaque progression and acute events. For example, Salminen and colleagues performed a genome-wide association study (GWAS) to that of plasma MMP-8 levels as well as overall cardiovascular events in a large cohort of patients. These investigators identified that in certain subsets of patients, polymorphisms within the MMP-8 promoter region were associated with different plasma MMP-8 levels. Past studies have identified that polymorphisms within the MMP-9 promoter region would also result in higher plasma MMP-9 levels and in turn impart increased risk for cardiovascular events. More importantly, however, Salminen and colleagues identified several important polymorphisms distal to that of MMP-8 itself using a GWAS approach: that of complement factor-H and in a specific member of the S100A family, both of which directly or indirectly may have influenced MMP-8 levels. While this is an oversimplification, these findings add to the body of evidence regarding an inter relationship between inflammatory signaling pathways, MMP induction, and cardiovascular risk.
Past studies of MMP polymorphisms and cardiovascular risk have identified both gender and ethnicity as important independent confounding variables in predictive cardiovascular risk models. Indeed, a past study reported that polymorphisms contained within the MMP-8 promoter region conferred increased risk for the progression of carotid artery disease, specifically in women. In another study, the severity of carotid artery disease was most strongly associated with a common MMP-9 polymorphism (-1562T) in women. With respect to ethnicity, specific MMP polymorphisms have been identified to confer increased risk of cardiovascular disease progression and cardiovascular events in Asians and in blacks of African descent. For example, a polymorphism within the MMP-9 promoter is more frequent in African-Americans and is associated with higher MMP-9 plasma levels. Since MMP-9 has been implicated to promote adverse cardiovascular remodeling, then this would imply a higher relative risk for disease progression with this MMP-9 polymorphism.
Specific MMP polymorphisms confer additive risk for cardiovascular events in subjects with other risk factors, such as diabetes and obesity. For example, MMP-9 polymorphisms identified in patients with metabolic syndrome appear to be synergistic with respect to cardiovascular events. Of potentially greater significance, MMP-2 and MMP-9 polymorphisms, which have been associated with cardiovascular disease progression ( see Table 4.1 ), have been more commonly identified in childhood obesity. While remaining speculative, these MMP polymorphisms may confer an increased risk for adverse cardiovascular remodeling at a much younger age. In a robust sample size of hypertensive patients (>42,000), Tanner and colleagues identified interactions between antihypertensive agents to that of MMP-9 and MMP-12 polymorphisms and subsequent cardiovascular events. This suggests that identifying specific MMP polymorphisms, particularly in more vulnerable patient populations, may provide additional guidance for specific pharmacotherapy. For example, MMP-1 polymorphisms have been shown to provide additive prognostic value in patients with acute coronary syndromes, and thus may identify a subset of patients for more intensive pharmacotherapy. Of course, simple sequence substitutions within the MMP gene or the promoter region provide only a partial picture into the complex regulation of MMPs within the cardiovascular system. Nevertheless, there is growing evidence that polymorphisms that affect MMP transcriptional control may play a critical role in how these proteases contribute to cardiovascular remodeling and, ultimately, cardiovascular outcomes.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here