Molecular, Gene, and Cellular Mechanism

Calcium Dysregulation

Calcium (Ca ²⁺ ) plays a central role in the excitation-contraction and repolarization-relaxation of the myocardium. Hence factors that regulate Ca ²⁺ flux in myocytes are poised to be critical regulators of diastolic function. Depolarization of the sarcolemma results in Ca ²⁺ influx into the cytosol via the voltage-gated L-type channels. This inward Ca ²⁺ current (I _Ca ) promotes the release of Ca ²⁺ from the sarcoplasmic reticulum (SR) by Ca ²⁺ –induced Ca ²⁺ –release via the ryanodine receptor 2 (RYR). The I _Ca and SR Ca ²⁺ release raise intracellular free Ca ²⁺ allowing Ca ²⁺ to bind to the myofilament protein troponin C (TnC). When cytosolic Ca ²⁺ is low, the troponin-tropomyosin complex inhibits the formation of the actinomyosin complex. When Ca ²⁺ binds to TnC, it releases the inhibition and allows cross-bridge cycling and contraction of the sarcomeres. In cardiac relaxation, cytosolic free Ca ²⁺ must decline to result in Ca ²⁺ dissociation from TnC. Ca ²⁺ is transported from the cytosol by four pathways: (1) SR Ca ²⁺ -ATPase (SERCA), (2) sarcolemmal Na/Ca exchanger (NCX), (3) sarcolemmal Ca ²⁺ -ATPase, and (4) mitochondrial Ca ²⁺ uniporter ( Fig. 1.1 ). For Ca ²⁺ homeostasis at each heart beat, the amount of Ca ²⁺ pumped back into the SR by SERCA must equal the amount released through the RYR2 channel, and levels of Ca ²⁺ extruded from the cell must equal the amount that entered via the L-type channel.

During relaxation Ca ²⁺ is pumped back into the SR lumen by SERCA, which is regulated by phospholamban (PLN). When PLN is dephosphorylated it binds to SERCA and inhibits its Ca ²⁺ affinity. Phosphorylation of PLN relieves SERCA inhibition and enhances Ca ²⁺ sequestration in the SR increasing the rate of relaxation. PLN is phosphorylated by cyclic adenosine monophosphate (cAMP) activated–protein kinase A (PKA) and Ca-CaM–dependent protein kinase (CaMK) as a result of β-adrenergic stimulation. The major substrates for the cAMP-PKA axis include PLN, L-type Ca channels, RYR, troponin I (TnI), and myosin-binding protein C (cMyBP-C). The relaxant effect of PKA is mediated mainly by phosphorylation of PLN and TnI. PLN phosphorylation (at Ser-16) speeds up SR Ca ²⁺ sequestration, while phosphorylation of TnI speeds up dissociation of Ca ²⁺ from the myofilaments. CaMKII phosphorylation of PLN (at Thr-17) also increases SR Ca ²⁺ -ATPase activity. Both PKA and CaMKII are likely to be coactivated during normal sympathetic stimulation (β adrenergic), creating synergy between these important regulatory signaling pathways (see Fig. 1.1 ). SERCA expression and function is decreased in most HF models. Several studies have also shown reduced SERCA/PLN ratio with age. In addition, there are data that point to reduced phosphorylation state of PLN in HF. This would result in reduced Ca ²⁺ sensitivity of SERCA and lower SR Ca ²⁺ uptake at physiologic cytoplasmic Ca ²⁺ levels [Ca] _I . The reduction of SERCA activity is consistent with the characteristic slowed relaxation and [Ca] _I seen in diastolic HF. In animal models when SERCA expression is increased or PLN expression is decreased, myocardial relaxation and [Ca] _I decline are accelerated resulting in improved diastolic function. Consequently, factors that might target SERCA and other mediators of Ca ²⁺ flux are an active area of research for therapeutic opportunities.

Posttranslational Modification of Myofibrillar Proteins

The sarcomeric protein titin is the largest known protein with a length greater than 1 um. It spans half the sarcomere connecting the Z-line to the M-line. It functions as a very large molecular spring contributing to force transmission at the Z-line and resting tension in the I-band region. Titin limits the range of sarcomere motion and is a major molecular contributor to myocyte passive stiffness. Titin activity can be modulated both by isoform expression and by phosphorylation. The differences in titin isoforms are correlated with the differences in the mechanical properties of cardiac, skeletal, and smooth muscle and differences of cardiac passive tension across species. Importantly, phosphorylation of titin by PKA, PKG, and PKC-α modulates its stiffness. Consequently, the activity of these kinases in myocytes has a direct influence on diastolic parameters. For example, PKA activated by β-adrenergic stimulation can phosphorylate titin as well as thick and thin filaments of the sarcomere. Nitric oxide (NO) and natriuretic peptides initiate signaling pathways activating PKG, which phosphorylates some of the same titin residues in the N2B spring element as PKA. Phosphorylation of the N2B element by either PKA or PKG results in a reduction in passive tension. α-Adrenergic stimulation activates PKC-α in cardiomyocytes, which is known to phosphorylate titin in the proline-valine-glutamate-lysine (PVEK) sequence increasing titin-based passive tension ( Fig. 1.2 ). Therefore phosphorylation of the N2B element in titin by PKA or PKG decreases passive tension, whereas phosphorylation of titin’s PEVK element by PKC-α increases passive tension. The increased oxidative pressure or ROS in diastolic dysfunction has been proposed to deplete NO reserve, lowering PKG activity and leading to hypophosphorylation of the N2B element and titin stiffing in HFpEF. Increased ROS can also result in the oxidation of cysteine residues in the N2B element resulting in disulfide bond formation and increased passive tension in mouse models.

In addition to titin, posttranslational modification of thick and thin filaments of myofibrils can affect cardiomyocyte relaxation. As mentioned, TnI and cMyBP-C are targets for phosphorylation by β-adrenergic stimulation. Phosphorylation of cardiac TnI at serine 23/24 by PKA reduces TnC-TnI interaction strength while reducing calcium sensitivity. The weakened C-I interaction may slow thin filament activation and result in faster relaxation kinetics thus increasing early phase relaxation with β-adrenergic stimulation. The cardiac cMyBP-C is a thick filament accessory protein that when unphosphorylated represses both cross-bridge attachment and detachment. It is phosphorylated by multiple kinases, including PKA, PKC, PKD, CaMKII, glycogen synthase kinase 3β, and ribosomal S6 kinase. Phosphorylation of cMyBP-C results in increased rates of cross-bridge cycling. Hypophosphorylation of cMyBP-C is associated with diastolic dysfunction in human patients. A recent study examined phosphorylation-deficient and phosphomimetic mutants of PKA-targeted cMyBP-C sites. They found that PKA phosphorylation of cMyBP-C threonine 35 results in accelerated cross-bridge detachment of myosin and actin, thereby enhancing relaxation.

Other Posttranslational Modifications

Advanced aging is associated with increased posttranslational modifications and has been associated with a systemic proinflammatory state (inflamm-aging) and development of HFpEF. Inflammation of the coronary microvascular endothelial cells leads to increased production of ROS. Oxidative stress can promote diastolic dysfunction by reducing the bioavailability of NO. Cardiac relaxation is regulated by NO. ROS can affect NO-related signaling at multiple sites. NO is generated by NO synthase (NOS), which requires tetrahydrobiopterin as a cofactor for the reaction. Hypertension and activation of the renin-angiotensin system lead to a depletion of tetrahydrobiopterin. The loss of tetrahydrobiopterin leads to NOS uncoupling, the production of superoxide instead of NO, and diastolic dysfunction. Some of the mechanisms of how NO modulates myofilament contractility have been recently revealed. Depletion of tetrahydrobiopterin in hypertension can repress NO synthesis and is associated with S-glutathionylation of MyBP-C, which reduces cross-bridge cycling. S-glutathionylation is an oxidative posttranslation modification of cysteines. Tetrahydrobiopterin supplementation lowers S-glutathionylation of MyBP-C, reduces the changes in actin-myosin cross-bridge cycling, and improves diastolic dysfunction. Further, excessive ROS leads to oxidation of guanylate cyclase and affects its responsiveness to NO to synthesize cyclic guanosine monophosphate (cGMP). Lower cGMP level decreases PKG activity in cardiomyocytes leading to hypophosphorylation of titin resulting in increased cardiac stiffness.

Many cardiac myofibrillar proteins are posttranslationally modified by acetylation in the healthy heart. Unfortunately, we know very little about the changes in acetylation with cardiac pathologies. One recent study demonstrated that histone deacetylase (HDAC) inhibitors were efficacious in two murine models of diastolic dysfunction. In addition, the investigators showed that HDAC2 copurified with myofibrils. Although the target(s) were not identified, the study showed that ex vivo deacetylation of isolated myofibrils with recombinant HDAC2 significantly increased the rate of myofibril relaxation, whereas acetylation with recombinant p300 decreased myofibril relaxation duration. Hence HDAC inhibitors might be a promising avenue for future research into potential HFpEF therapies.

Transthyretin Amyloidosis

Cardiac amyloid deposition has also been linked with HFpEF. Over 30 different proteins have been shown to form amyloid fibrils and five (immunoglobulin light chain, immunoglobulin heavy chain, transthyretin, serum amyloid A, and apolipoprotein AI) have been found to infiltrate the heart. Autopsy from patients who were diagnosed with HFpEF revealed that transthyretin amyloidosis was present in 32% of those greater than 75 years of age. Another study using nuclear scintigraphy to detect amyloids has indicated that 13% of hospitalized patients with HFpEF have transthyretin amyloidosis. Interestingly, some HF patients have amyloidosis caused by a mutation in transthyretin. Over 80 transthyretin mutations, with autosomal dominant inheritance, have been associated with tissue amyloid deposition, some within the heart. Nearly 25% of African Americans with cardiac transthyretin amyloidosis were heterozygous for a transthyretin V122I mutation. Although this condition is rare, interstitial deposition of wild-type and mutant transthyretin is an underrecognized trigger of HF in the elderly.

Mitochondrial Dysfunction and Age

As evidenced in the preceding sections, there are many factors that contribute to cellular changes observed in HFpEF. However, the fact that aging is a critical and overarching factor in the development of HFpEF is clearly noted. LV diastolic stiffness increases and LV diastolic filling rate decreases with age. Kaushik et al. showed that age-related increases in vinculin, a cytoskeletal protein, was linked to cortical stiffening and contractility. Vinculin is found localized to integrin-mediated cell–ECM and cadherin-mediated cell–cell adhesions. Vinculin acts as one of several proteins involved in anchoring F-actin to the membrane. Senescent rats have twofold the ECM content in the myocardium compared to younger rats. The senescent myocardium has increased levels of ROS, which can activate transforming growth factor-β (TGF-β), inducing conversion of cardiac fibroblast to myofibroblasts, an activated fibroblast phenotype. The aging heart has lower responsiveness to β-adrenergic stimulation. There is lowered PKA and CaMKII phosphorylation of PLN and RYR receptor. Together this results in a lowering of the Ca ²⁺ uptake and relaxation rate.

The mitochondrial deoxyribonucleic acid (DNA) in aging hearts in both man and mice have up to 16-fold more point mutations and deletions than those of younger animals. Myocardial energetics have been examined as a potential mechanism for reduced systolic reserve in HFpEF with increased age. Patients with HFpEF have been shown to have a reduced phosphocreatine/adenosine triphosphate (ATP) ratio when compared to controls. Several studies suggest that abnormal skeletal muscle performance is a contributor to exertional intolerance rather than just limited cardiac reserve. One study found that HFpEF patients had reduced type I oxidative muscle fibers, type I/II fiber ratio, and a reduced capillary to fiber ratio in skeletal muscle compared with controls.

Comorbid diseases, including hypertension and renal failure, are much more common in the elderly. There is increased inflammation with increasing age. Although the cellular mechanisms are not yet clearly defined, myocardial aging is interconnected to molecular events that influence both myocyte contractility and changes in the collagenous ECM (see upcoming discussion) that appear to provide a favorable milieu for the development of HFpEF, particularly when in combination with other comorbitidies such as hypertension and diabetes.