Administration of a TLR9 Inhibitor Attenuates the Development and Progression of Heart Failure in Mice

Visual Abstract

Previous extensive studies on heart failure have reported an important role for proinflammatory cytokines in its pathogenesis (1). Circulating levels of proinflammatory cytokines, including tumor necrosis factor (TNF)-a, are related to the severity and prognosis of the disease. However, the targeted anti-TNF-a approaches were neutral with respect to the primary endpoints of the trial or resulted in worsening heart failure or death (2,3). In addition to TNF-a, the pro-inflammatory cytokines that are elaborated in heart failure include other members of the TNF superfamily, members of the interleukin-1 family, and interleukin (IL)-6 (1). Recognizing the molecular mechanism underlying the developing inflammation in heart failure is essential for developing strategies to control disease progression, including therapeutic drugs.
Mitochondrial deoxyribonucleic acid (DNA) contains the unmethylated cytidine-phosphateguanosine (CpG) motif, which stimulates Toll-like receptor (TLR) 9 to induce inflammation (4,5). Mitochondria damaged by external hemodynamic stress are degraded by the autophagy/lysosome system in cardiomyocytes (6). Insufficient degradation of mitochondrial DNA mediated through autophagy in pressure-overloaded mouse hearts leads to its binding to TLR9 to induce inflammation and heart failure (7). In failing mouse hearts, mitochondrial DNA is located in autolysosomes. Furthermore, TLR9 ablation in pressure-overloaded mice attenuated the development of inflammation and heart failure.
Thus, interference with TLR9 function by small molecules is likely to produce a better clinical outcome by preventing its aberrant inflammatory responses. [d]oxazole), is a synthetic antagonist of nucleic acid-sensing TLRs and is orally bioactive (8,9). In vitro, E6446 specifically inhibits the activation of TLR9 (8). Others have reported that the compound inhibits TLR9 but also TLR7 in a ligand-dependent manner (9). When E6446 is administered to mice, it suppresses inflammatory responses to challenge doses of unmethylated CpG containing oligodeoxynucleotide (CpG ODN) (8,9). When E6446 is administered chronically in mouse cerebral malaria and spontaneous lupus models, the compound inhibits cytokine production with prevention of signs of cerebral malaria and circulating antinuclear antibodies, respectively.
In the present study, the efficacy of oral treatment with E6446 was evaluated on mouse pressure overload-induced heart failure models. Our results indicate that E6446 exerts beneficial effects for the prevention and treatment of heart failure in mice and might be a novel therapeutic agent for treating patients with heart failure.

METHODS
CELL CULTURE. Adult cardiomyocytes were isolated from 10-to 12-week-old C57BL/6J male mice (CLEA Japan, Inc., Tokyo, Japan) by using a Langendorff system and were then cultured (7).    A blood sample was taken from the right ventricle. HISTOLOGICAL ANALYSIS. Heart samples were fixed in buffered 4% paraformaldehyde solution and embedded in paraffin (10). Fibrosis fraction was quantified by using ImageJ software (National Institutes of Health, Bethesda, Maryland) and  The time dependence of (A) cytokine protein and (B) messenger ribonucleic acid (mRNA) expression in the heart after injection with ODN1668. Mice (body weight 24.4 to 25.8 g) were pretreated with oral administration of 1.5 mg/mouse (60 mg/kg) of E6446 1, 2, or 3 days before intraperitoneal injection with 60 mg/mouse of ODN1668. Two hours after ODN1668 injection, mice were sacrificed for analysis (see Figure 9A). Dose dependency in the inhibition of (C) cytokine protein and (D) mRNA expression after ODN1668 injection with increasing concentrations of E6446. Two days after oral administration with the indicated dose of E6446, mice were administered an intraperitoneal injection of 60 mg/mouse of ODN1668. Two hours later, mice were sacrificed for analysis (see Figure 9B).  The treatment with E6446 was performed every 2 days from 2 days before TAC. The mice were analyzed 4 weeks after the operation (see Figure 10A   Two weeks after TAC, treatment with E6446 was performed every 2 days (see Figure 10B). Values are mean AE SE (n ¼ 10).  Figure 3A). When E6446 was administered 2 or 1 day before ODN1668 injection, IL-6 levels were lower than the E6446-nontreated ODN1668injected group and showed no significant difference compared with the control group without ODN1668 injection. When E6446 was administered 3 days before ODN1668 injection, the level of IL-6 did not differ from that in the nontreated group. To confirm the protein data, mRNA levels in the heart were measured ( Figure 3B).  Values are mean AE SE. Thirty mice were subjected to TAC operation for 2 weeks. Ten mice with fractional shortening >50% were excluded from the study. The remaining 20 mice were randomized to the saline-and E6446-treated groups. The parameters of the mice were obtained 2 weeks after the operation by using echocardiography. *p < 0.05 vs. corresponding control at baseline. Table 1. To label mitochondrial DNA, mice were injected with EdU 1 day before sacrifice. EdU specifically binds to mitochondrial DNA during active DNA synthesis in nondividing cardiomyocytes (7). LAMP2a is a marker for lysosomes. In TAC-operated saline-and ined. Mice were subjected to TAC operation ( Figure   8A) and divided into 2 groups 2 weeks later. There were no significant differences in echocardiographic parameters between the 2 groups, which already exhibited chamber dilatation and cardiac dysfunction ( Table 2). The mice were then administered E6446 or saline every 2 days for 4 weeks. LV chamber dilatation and cardiac dysfunction progressively worsened with time in both groups ( Figures 8B and 8C). End-diastolic LV internal dimensions, end-systolic LV internal dimensions, and LV mass showed no significant difference between saline-and E6446-treated TACoperated mice until 3 weeks after TAC. However, the parameters were significantly smaller in E6446treated mice than those in saline-treated mice 4 and 6 weeks after the operation. Fractional shortening was significantly higher in E6446-treated mice than that in saline-treated mice 6 weeks after TAC. There were no significant differences in IVSd and end-diastolic LV posterior wall thickness between the 2 groups at any time point. The heart weight-to-tibia length and lung weight-to-tibia length ratios were significantly reduced by E6446 treatment ( Figure 8D). Schematic protocols to examine the effect of treatment of E6446 on cardiac phenotypes are described in Figures 9 and 10.

DISCUSSION
The present study showed that E6446 prevents the development of pressure overload-induced heart failure when administered before the cardiac event and also suppresses the progression of heart failure when started after cardiac dysfunction manifested.
We have reported that TLR9 is essential in producing inflammatory cytokines in failing hearts (7). TLRs are essential in driving the recruitment of inflammatory cells and production of cytokines during cardiac remodeling (13). E6446 prevents cellular events acti-  (A) Schematic protocol to examine the effect of treatment with E6446 initiated before TAC on cardiac phenotypes. E6446 (1.5 mg/mouse) (n ¼ 9) or saline (n ¼ 9) was administered every 2 days from 2 days before TAC. Blood pressure was measured 7 days after TAC. Four weeks after TAC, mice were subjected to echocardiographic analysis and sacrificed. (B) Schematic protocol to examine the effect of treatment with E6446 initiated after TAC on cardiac phenotypes. Thirty mice were subjected to TAC operation for 2 weeks. Ten mice with fractional shortening >50% were excluded from the study.
The remaining 20 mice were randomized to the saline-and E6446-treated groups and then administered saline or E6446 (1.5 mg/mouse) every 2 days. Six weeks after TAC, mice were subjected to echocardiographic analysis and sacrificed. Abbreviations as in inhibited the expression of inflammatory cytokines through a TLR9-dependent pathway but not TLR4or TLR7-dependent pathways in adult cardiomyocytes. Thus, E6446 has high specificity to TLR9. In mouse bone marrow-derived dendritic cells, E6446 potently inhibited IL-6 production induced by CpG ODN but not by TLR3 ligands (9).
However, E6446 was a potent inhibitor of IL-6 induction by single-stranded RNA, a TLR7/8 agonist, but a relatively poor inhibitor of IL-6 induction by R848, suggesting that the ability of E6446 to suppress TLR7/8 might be ligand dependent. Based on our data showing the importance of TLR9 signaling in the development of inflammation and heart failure and its specificity to TLR9 in cardiomyocytes, the cardioprotective action of E6446 is TLR9 mediated. However, we cannot exclude the possibility that TLR7/8 is involved in the effect of E6446 on the development of heart failure.
E6446 inhibits in vitro DNAÀTLR9 interaction via an association with DNA but not with TLR9 (9). Furthermore, E6446 accumulates in the intracellular acidic compartment. Mitochondrial DNA is accumulated in autolysosome and coexists with TLR9 in failing hearts (7). DNase II activity was up-regulated in hypertrophied hearts but not in failing hearts. The incomplete digestion of mitochondrial DNA would be due to the loss of up-regulation of DNase II activity. Mitophagy impairment occurs in the TAC-induced mouse heart failure model (16). Thus, it is also possible that impairment of mitophagy or lysosomal permeabilization or lysosomal dysfunction might result in the accumulation of mitochondrial DNA in autolysosome.
Our data in this study showed that there was no significant difference in the number of EdU and LAMP2a double-positive deposits between TAC-operated saline-and E6446-treated hearts. This outcome suggests that E6446 has no effect on mitochondrial DNA accumulation in autolysosomes. Thus, we can assume that the E6446 administered accumulates in lysosomes in the cardiomyocytes and interacts with mitochondrial DNA. When E6446 was orally administered to mice before TAC, E6446 inhibited TLR9 signaling by interfering with the mitochondrial DNAÀTLR9 interaction and subsequent development of inflammation and heart failure.
We showed that Tlr9 -/mice are more resistant to pressure overload than control mice, and inhibitory ODN to TLR9 (ODN2088) improved survival in TACoperated wild-type mice when administered before TAC (7). However, administration of a drug before cardiac events is not clinically relevant. The results indicate that E6446 can slow the development of heart failure even after cardiac dysfunction manifested. Thus, E6446 or other immunomodulatory therapy can be used to prevent or delay pressure overload-induced heart failure.
STUDY LIMITATIONS. This study shows the therapeutic effects of a TLR9 inhibitor, E6446, on mouse pressure-overload heart failure model. Obviously, further studies are necessary to translate the findings into human heart failure therapy. TAC-induced mouse model does not fully represent the complex features of clinical heart failure. To establish the clinical feasibility of E6446 treatment for heart failure, the effects of E6446 on different heart failure models have to be examined, such as myocardial infarction. In addition, we used young healthy mice in this study. However, in most patients, and particularly in elderly patients,

CONCLUSIONS
Heart failure is the result of various cardiac diseases such as myocardial infarction, high blood pressure, cardiomyopathy, valvular diseases, arrhythmia, and congenital heart diseases. Elevated levels of inflammatory mediators have been identified in patients with heart failure, including heart failure with reduced and preserved ejection fraction, as well as short-term decompensated heart failure (1). Thus, investigation of the involvement of the TLR9signaling pathway in other mouse or larger animal heart failure models and various types of human heart failure is warranted. We ultimately will be able to identify subsets of patients with heart failure who will benefit from inhibition of TLR9 signaling.