Cholinergic Elicitation Prevents Ventricular Remodeling via Alleviations of Myocardial Mitochondrial Injury Linked to Inflammation in Ischemia-Induced Chronic Heart Failure Rats

Background Cholinergic anti-inflammatory pathway (CAP) is implicated in cardioprotection in chronic heart failure (CHF) by downregulating inflammation response. Mitochondrial injuries play an important role in ventricular remodeling of the CHF process. Herein, we aim to investigate whether CAP elicitation prevents ventricular remodeling in CHF by protecting myocardial mitochondrial injuries and its underlying mechanisms. Methods and Results CHF models were established by ligation of anterior descending artery for 5 weeks. Postoperative survival rats were assigned into 5 groups: the sham group (sham, n = 10), CHF group (CHF, n = 11), Vag group (CHF+vagotomy, n = 10), PNU group (CHF+PNU-282987 for 4 weeks, n = 11), and Vag+PNU group (CHF+vagotomy+PNU-282987 for 4 weeks, n = 10). The antiventricular remodeling effect of cholinergic elicitation was evaluated in vivo, and H9C2 cells were selected for the TNF-α gradient stimulation experiment in vitro. In vivo, CAP agitated by PNU-282987 alleviated the left ventricular dysfunction and inhibited the energy metabolism remodeling. Further, cholinergic elicitation increased myocardium ATP levels and reduced systemic inflammation. CAP induction alleviates macrophage infiltration and cardiac fibrosis, of which the effect is counteracted by vagotomy. Myocardial mitochondrial injuries were ameliorated by CAP activation, including the reserved ultrastructural integrity, declining ROS overload, reduced myocardial apoptosis, and enhanced mitochondrial fusion. In vitro, TNF-α intervention significantly exacerbated the mitochondrial damage in H9C2 cells. Conclusion CAP elicitation effectively improves ischemic ventricular remodeling by suppressing systemic and cardiac inflammatory response, attenuating cardiac fibrosis and potentially alleviating the mitochondrial dysfunction linked to hyperinflammation reaction.


Introduction
Chronic heart failure (CHF), one of the most lifethreatening public health problems, has resulted in many global economic and living burdens. Ischemic events such as myocardial infarction (MI) have been acknowledged as the most common etiology of CHF [1]. Due to its intricate pathogenesis, over 30% of patients with normalized pharmacological therapies or mechanotherapy still experience deteriorations in cardiac function and poor prognosis during long-term follow-up [2,3]. Hence, it is meaningful to explore additional available therapeutic strategies.
CHF secondary to the ischemia generally results in adverse cardiac remodeling [4]. Current knowledge suggests that a sustained inflammatory response resulting from the initial myocardial injury, as well as the maladaptive cardiac energy metabolism due to mitochondrial dysfunction, jointly participate in decompensated ventricular remodeling [5][6][7]. Notably, chronic inflammation mediated by immune cells or proinflammatory cytokines has a close association with mitochondrial dysfunction [8]. For instance, a recent study reported that tumor necrosis factor α (TNF-α) preconditioning directly induced mitochondrial fragmentation and increased membrane potential and reactive oxygen species (ROS) overproduction in cardiomyocytes [9]. Moreover, mitochondrial-related damage accelerates inflammatory response in turn by activating the NLR family pyrin domain containing 3 (NLRP3) inflammasomes and downstream regulators of the immune signaling pathway like toll-like receptors (TLRs) [10,11]. In short, the detrimental crosstalk between myocardial mitochondrial injury and hyperinflammation may further establish a damaging pathophysiological cycle and facilitate the ventricular remodeling [12].
The cholinergic anti-inflammatory pathway (CAP) is an endogenous self-protective mechanism connecting the autonomic neural reflex to immune regulation. Activating alpha 7 nicotinic acetylcholine receptors (α7nAChRs) expressed on macrophages or monocytes mediates a decrease in the transcription of downstream proinflammatory cytokines in the context of an intact vagal nerve [13][14][15]. Emerging evidence showed that cholinergic stimulation was able to improve cardiac remodeling and hemodynamics in the postinfarction rats, and this effect was reversed by the peripheral blockade of α7nAChR [16]. Other studies demonstrated the beneficial effect of activating the CAP on the arrhythmogenic substrate with ischemic cardiomyopathy, and the underlying mechanisms rely on the inflammation resolution and cardiac fibrosis attenuation [17,18]. There are no ongoing studies regarding whether CAP elicitation ameliorates the myocardial mitochondrial injuries in the postinfarction state. Here, we aimed to study the following: (1) to verify the effect of cholinergic stimulation on ventricular remodeling especially in the cardiac energy metabolism state, in CHF models; (2) to ascertain the role of CAP activation in mitochondrial dysfunction in CHF; and (3) to uncover the potential mechanisms related to the cardioprotective effect of stimulating CAP.

Surgical Procedures
2.3.1. MI Establishment. All operations were carried out in a laminar flow room under aseptic conditions. Before all the invasive manipulations, each rat was generally administered 1 ml of sodium pentobarbital (2 mg/100 ml) to induce anesthesia and atropine (0.05 mg/kg) to prevent airway secretions. A multichannel physiology recorder (Lead-7000 EP CONTROL, Sichuan, China) was used to monitor the intraoperative heart rate. The anesthetized rats were fixed on a bubble brick and placed in the right lateral position. A cannula with an outer diameter of 2.5 mm (Jinkou 2.5 mm, Yu Yan Instruments, Shanghai), which was connected to the animal respirator (Ugo Basile 5025), was used for tracheal intubation with a tidal volume of 8 ml/kg. The left ventricle was exposed after thoracotomy at the left third intercostal space. The left anterior descending (LAD) branch was ligated approximately 2-3 mm below the left atrial appendage (LAA) with a 7-0 polyglactin suture. The successful MI model was determined by the ST-segment alterations from the II-lead of the ECG during a postoperative half hour. After the surgery, 50 mg of pethidine and 400 U of penicillin were administered intramuscularly for analgesia and infection prevention, respectively [19].

Vagotomy.
We subjected CHF rats to the right cervical vagotomy during anesthesia in the fifth week. The skin was incised along the right cervical line before the blunt dissection of the subcutaneous muscle layer. The right cervical vagal nerve was located in the carotid sheath, meticulously isolated by a glass needle, and cut off with ceramic scissors [20]. After the vagotomy, the incised skin was closed with interrupted sutures, and the rats were individually housed.
2.4. Echocardiography. Transthoracic echocardiography was performed weekly beginning in the first postinfarction week and lasting until the fifth week. Echocardiographic examination was implemented with a Philips HD11XE transthoracic Doppler ultrasound imaging system (Philips Inc., Bothell, WA, USA) as described in our previous work [21]. The 2.6. Transmission Electron Microscopy (TEM). The animals were euthanized by overdose anesthesia at the end of this experiment. Tissues were collected from the peripheral infarction area of the LV lateral wall. Each specimen was trimmed to approximately 1 mm 3 size and directly immersed in precooled 2.5% glutaraldehyde at 4°C overnight (Sbjbio, China, SBJ-0639M). Then, the fixed specimens were subjected to gradient dehydration in an acetone solution, uranium acetate staining, and lead citrate staining as described in our previous study [22]. The ultrastructure of cardiomyocytes was observed by a JEM-100 CXII TEM (JEOL, Japan, 80 kV), and images were captured at 500x magnification.

Adenosine Triphosphate (ATP) Concentration
Determination. Fresh tissue samples were prepared, homogenized, and centrifuged to extract the supernatant (3000 r/ min, 15 minutes). The LV myocardium ATP levels were examined by an ATP assay kit (Nanjing Jiancheng Bioengineering Institute, China, A095-1-1) according to the product instructions. OD values of the tested samples were measured at a wavelength of 636 nm, and ATP concentrations were calculated based on the standard curve.
2.8. Histology Study. LV tissues were cut into small pieces, fixed with 4% paraformaldehyde, and embedded in paraffin. Hematoxylin-eosin (HE) staining, Masson's trichrome staining, and picrosirius red staining were performed on the 5 μm thick histological sections to analyze LV fibrosis in the light of our previous method [23]. Eight paraffin blocks per group were taken for the tissue slide preparation and 9 separate bright field microscopic images per animal were quantified. The degree of fibrosis was analyzed by ImageJ software version 1.8.0. The fibrosis size was defined as the proportion of the collagen-positive area in the total myocardial area.
2.9. Immunohistochemistry. Tissue paraffin blocks of different groups were selected to be sliced and subjected to specimen preparation, antigen repair, and background sealing. Primary antibodies against CD68 (ab125212, Abcam, Cambridge, UK) and CD163 (ab182422, Abcam, Cambridge, UK) were added to the tissue sections and detected by the goat anti-rabbit IgG (H&L) secondary antibodies (ab150077, Abcam, Cambridge, UK). Similar to the quantitative method in histology, immunoreactivity was expressed as the entire positive staining area per square millimeter and was calculated manually by ImageJ software version 1.8.0.

Terminal Deoxynucleotidyl Transferase dUTP Nick End
Labeling (TUNEL) Staining. LV peri-infarct myocardial tissue was embedded in paraffin and routinely sliced into 5 μm sections for TUNEL staining. After a series of processes including deparaffinization, rehydration, and cell transparency, 50 μl of the TUNEL mixture was added to the sections according to the instructions for the kit (No.11684817910, Roche, USA). Subsequently, the sections were incubated with converter pods for 30 minutes at 37°C, and DAB color reagent was used for chromogenic detection. Analysis of the apoptosis rate, which reflected the myocardial apoptosis level, was performed by selecting eight slides in each group randomly and then taking photos manually by laser scanning confocal microscopy (LSCM). The proportion of apoptotic cardiomyocytes was also measured using Image-Pro Plus software version 6.0.
2.11. Cell Culture and Treatment. H9C2(2-1) cells (Cell Bank, Chinese Academy of Sciences) were incubated in 90% DMEM medium (12800017, GIBCO) containing NaHCO 3 (1.5 g/l) and supplemented with 10% high-quality FBS (HN-FBS-50, HAKATA). All cell lines were cultured in a 5% CO 2 atmosphere at 37°C. Cell viability was measured by a cell growth curve, and cell proliferation was evaluated by the CCK-8 growth curve. Different concentrations of TNF-α (0, 5, 10, 20, 30, and 50 ng/ml) were used for choosing the optimal concentration in vitro treatment of cell lines. Untreated H9C2 cells were added as a comparison control group.

Mitochondrial Permeability Transition Pore (mPTP)
Measurement. The cells were lysed in a lysis buffer, preliminarily transferred, and centrifuged. Then, we added 1 ml of calcein AM staining solution to 10 ml of assay buffer, and the mixture was used to resuspend the cells. Moreover, a tube of cells in 1 ml of PBS was used as the negative control group. The single-cell suspension was prepared as previously mentioned. mPTP assessments were also achieved by flow cytometry.
2.12.4. Apoptosis Evaluation. The acquisition and collection of cells from tissue and cell lines were similar to the procedures described for measuring ROS. The single-cell suspension was prepared by adding 500 μl of 1x binding buffer to resuspend the cells, which were passed through a 200-filter mesh. Five microliters of Annexin V-PE and 10 μl of 7-AAD were added to each tube. After gentle mixing and 10 minutes of rest at 4°C, flow cytometry was performed within 30 minutes.
2.13. Western Blotting. Excised ventricular tissues were promptly transferred to liquid nitrogen, and tissue homogenate was prepared. Tissue and cell samples were lysed with 200 μl of RIPA lysis buffer with protease inhibitors and mixed adequately on a shaker for 1 hour. The supernatant was collected after high-speed centrifugation for 15 minutes at 15000 rpm. Protein concentrations were determined by a BCA assay kit (23235, Thermo Fisher). The quantified protein was mixed with 5x loading buffer and boiled for 5 minutes at 100°C. Electrophoresis was performed using a 10% SDS-PAGE separation gel. Semidry transfer was carried out on the PVDF membrane for 60 minutes at 100 V. Then, 5% BSA powder sealant was used to seal the membrane. After incubating overnight at 4°C with primary antibodies, we incubated the membrane with diluted secondary antibodies at room temperature for 1 hour. Finally, the chemiluminescence reagent was selected for coloration. H9C2 cells were rapidly scraped from the culture dish surface rapidly. 2.14. Statistical Analysis. All raw data were input into Excel version 2007 and were analyzed by SPSS version 26.0. Continuous variables between the control and the TNF-α group are described by means ± standard deviations and were analyzed by Student's t -tests. Univariate ANOVA with Bonferroni post hoc comparisons was used to analyze the differences in cardiac function parameters, circulating marker levels, local immune infiltration degrees, fibrosis   Mediators of Inflammation and apoptosis extents, and protein expression levels among the five groups. Correlations between LVEDd and the levels of IL-1β, IL-6, TNF-α, and ATP levels were analyzed by Pearson's correlation coefficient. Statistical significance was defined as a 2-tailed P value < 0.05.

Cholinergic Elicitation Restored Cardiac Dysfunction and
Myocardial Energy Metabolism Alterations. All experimental procedures were conducted according to the study protocol (Figure 1(a)). In this study, the postoperative mortality of rats was approximately 30% (Figure 1(b)). Typical M-mode ultrasound schematic diagrams are shown in Figure 2(a). For echocardiographic data during the 9 th week, the LVEDd was enlarged in the CHF group in comparison to that in the sham group, and this parameter was further increased in the Vag group while decreased in the PNU group (P < 0:05, Figure 2(b)). The LVEF and LVFS values obviously declined in the Vag group. Compared to those in the CHF group, both the LVEF and LVFS values were larger in the sham and PNU groups (P < 0:05, Figures 2(c) and 2(d)). As shown in the Vag+PNU group, vagotomy counteracted the protective effect of PNU on LVFS but not LVEDd or LVEF (P < 0:05, Figure 2(d)). The exact values of the echocardiographic parameters are displayed in Supplemental Table 1. The decline in intramyocardial ATP contents in the CHF state was prevented by PNU treatment and was exacerbated by vagal mutilation (P < 0:05, Figure 2(f)). Correlation analysis showed that the ATP level negatively moderately correlated with LVEDd (r = −0:609, P < 0:05, Figure 2(g)). The changes in BNP, INS, NEFA, and Glu levels in the five groups showed a consistent trend with LVEDd from the 6 th week to the 9 th week, which was enhanced by vagotomy and weakened by PNU administration. Until the 9 th week, compared with those in the PNU group, four indicator values in the Vag+PNU group increased (Figure 2(e) and Figures 2(h)-2(j)).

Cholinergic Elicitation Inhibited Systemic Inflammation and Cardiac Macrophage Infiltration in the Postinfarct
Phase. The levels of three cytokines (IL-1β, IL-6, and TNF-α) that participate in the systemic inflammatory response are shown in Figures 3(e)-3(i). As expected, the plasma levels of these three proinflammatory factors in plasma were lowered by one week of PUN administration but were increased by vagotomy, and this trend persisted until the 9 th week. Pearson's correlation analysis showed the moderate positive correlations between IL-1β, IL-6, and TNF-α and LVEDd (IL-1β: r = 0:691, P < 0:05; IL-6: r = 0:717, P < 0:05; TNF-α: r = 0:824, P < 0:05, Figures 3(f)-3(j)). Immunohistochemical staining was used to evaluate the abundance of local macrophages in the myocardium. As presented in Figures 3(a) and 3(b), CD68 + and CD163 + macrophages, which represented the M1-like and M2-like phenotypes, respectively, were mostly found in the peri-infarction border areas at the 9 th week post-MI. Quantitative analysis demonstrated that the number of CD68 + macrophages increased in the Vag group and decreased in the PNU group, whereas the numbers of CD163 + macrophages peaked in both the Vag and PNU groups (P < 0:05). The impact of PNU cure on macrophage infiltration was counteracted by vagotomy to some extent (P < 0:05, Figures 3(c) and 3(d)).  Figure S1A, some cardiomyocytes were covered by the accumulated inflammatory cells in a cross-sectional view. The vast formation of connective tissue in the perivascular space is presented in Figure S1B. In Figure S1C, a mass of fibrous tissue hyperplasia was accompanied by few inflammatory cells distributed around the myocardial interstitial substance. Figure S1D shows that the disrupted cardiomyocytes were replaced by the extensive production and deposition of fibrous tissue. HE staining shows scar regions in the five groups (Figure 4(a)). The Masson's trichrome staining and picrosirius red staining were performed to identify the differences in collagen deposition among the five groups. The

Mediators of Inflammation
Masson's trichrome staining showed that there were no significant differences in the total collagen areas between the CHF and Vag groups (P < 0:05, Figure 4(b)). However, as shown in the picrosirius red-stained sections, chronic infarction/ischemia induced collagen accumulation, which were reduced by PNU administration and were augmented by vagotomy (P < 0:05, Figure 4(c)).

Cholinergic Elicitation Alleviated
Mitochondrial-Dependent Apoptosis. TUNEL staining was used to assess myocardial apoptosis. As shown in Figure 5(a), the number of brown DIG-d-UTP-stained brown cells, which represented apoptotic cells, was markedly increased in CHF and Vag groups and decreased in the PNU group (P < 0:05, Figures 5(a) and 5(b)). Three proteins related to the mitochondrial apoptotic pathway were tested by western blotting. Quantitative analysis showed that the expression levels of caspase-3 and Bax were elevated in the CHF phase and were further upregulated by vagus transection and downregulated by PNU treatments. In contrast, the relative expression of Bcl-2 was reduced in the Vag group and augmented in the PNU group (P < 0:05, respectively, Figures 5(c)-5(g)).  Figure 6. In the sham group, the well-defined mitochondria and intact sarcomeres were arranged regularly, and there were clear Z-lines. The abundant mitochondria are shown in the detailed panel. However, a panel of the CHF group showed the damaged mitochondria scattered among snatchy Z-lines tangled with the disorganized myofibers. In the CHF group, most mitochondria were present with dissociated myofibers near a few distinct lipid droplets. Similarly, a large area of impaired mitochondria accompanied by broken fibers was present in the CHF+Vag group. Furthermore, a reduction in mitochondria, the disappearance of inner and outer membranes, the absence of cristae, and mitochondrial vacuolization were clearly observed. In the CHF+PNU group, lipid droplets were surrounded by lysosomes, and most mitochondria maintained regular morphologies with slightly swollen mitochondrial cristae. Finally, the CHF+Vag+PNU group exhibited distorted cristae and discontinuous membranes.  Figure 7(a). Quantitative analysis demonstrated that ROS levels increased in the Vag group and decreased in the PNU group (P < 0:05, Figure 7(b)). PNU treatment significantly downregulated mitochondrial fission-related proteins, such as Fis1, and upregulated fusion-related proteins as Mfn1 and Opa1, in comparison to those postvagotomy. However, the expression levels of other mitochondrial fission-related protein as Drp1 and fusion-related protein as Mfn2 showed insignificant differences among five groups (P > 0:05, Figures 8(a)-8(f)). The expression of membrane receptor α7nAchR was downregulated by vagotomy and was partially counteracted by PNU administration (P < 0:05, Figure 8(g)). Besides, vagotomy obviously lowered the expression levels of phosphorylated transcription factors NF-κB and STAT3, which were rescued by PNU treatment (P < 0:05, Figures 8(i) and 8(m)). On the contrary, the protein levels of both NF-κB, JAK2, phosphorylated JAK2, and STAT3 were not altered among five groups (P > 0:05, Figures 8(h) and 8(j)-8(l)).

TNF-α Mediated Mitochondrial Injury in Cardiomyocytes in a Concentration-Dependent Manner.
A TNF-α gradient experiment was performed to determine a target concentration for in vitro experiments. Growth curve of H9C2 cells was depicted in Figure 9(a). With increasing concentrations of TNF-α from 0 ng/ml to 50 ng/ml, intercellular ROS production increased (Figure 9(b)). TNF-α at 50 ng/ml was chosen for the optimal intervention concentration in vitro study. As anticipated, compared with the control group, TNF-α induced cardiomyocyte apoptosis and facilitated the reduction in MMP and opening of the mPTP without affecting cell proliferations (Figure 9(c), P > 0:05; Figures 9(d)-9(f), P < 0:05, respectively). Consistent with the western blot results at the tissue level, TNF-α stimulation significantly inhibited mitochondrial fusion-related markers such as Mfn1 and Opa1 (P < 0:05, Figure 9(g)).

Discussion
The major findings of the present study were as follows: (1) CAP activation ameliorated the impaired cardiac function, reduced the LV dilatation, and restored abnormal Glu and lipid metabolism in the chronic infarction phase; (2) CAP  Figure 10). Clinically, CHF challenged by MI tends to be a consequence of cardiac pathologic remodeling during the postinfarction late phase [24]. First, in order to simulate the pathophysiological features of ischemic HF, we established the postinfarction CHF models, which successfully mim-icked the hallmarks including the altered ventricular structure, the impaired pump function, and disordered energy metabolism after a five-week cardiac remodeling period [25,26].
Then, in order to examine the macroscopic effect of CAP modulation on cardiac dysfunction and energy metabolism state, we bidirectionally manipulated CAP activities with PNU treatment and vagotomy in the 6 th week. Coinciding with previous results regarding the salutary effect of cholinergic induction on cardiac dysfunction [27][28][29], we demonstrated that activating the CAP improved ventricular structural remodeling and mechanical function, as evidenced by echocardiographic parameters and BNP levels.

Mediators of Inflammation
For the first time, to our knowledge, we reported that the levels of INS, NEFAs, and Glu continuously increased in CHF rats, which was partly inhibited by CAP stimulation and exacerbated by vagotomy. To explain this result, physiological evidence suggested that enhanced lipolysis induced by hypersympathetic activities worsened the reduction in Glu uptake and increased insulin resistance in advanced HF [30]. It was verified that cholinergic elicitation can modulate autonomic nerve activity and especially suppress sympathetic activities in many cases [31,32]. In addition, we found that insufficient ATP levels were moderately correlated with LV expansion, and an improved energy supply was achieved by targeting the CAP. From these observations, we hypothesized that mitochondria, which link substrate utilization and ATP production, may be essential for the protective effect of cholinergic stimulation.
Next, we focused on the three key pathophysiological responses to HF progress (inflammation, fibrosis, and mitochondrial injury) to describe the detailed beneficial effects of cholinergic elicitation. Solid evidence suggests that a prolonged inflammatory response challenged by myocardial injury, namely, chronic inflammation, is implicated in pathological cardiac remodeling and also has the crosstalk with cardiac fibrosis and mitochondrial injuries [33][34][35]. In our study, the systemic inflammatory reaction in the postinfarction stage was mitigated by activating the CAP but intensified by unilateral vagotomy, as proven by serum proinflammatory cytokine levels. The levels of IL-1β, IL-6, and TNF-α changed from the first week after intervention, which suggested a rapid antiinflammatory effect of cholinergic stimulation. However, circulatory cytokine levels are susceptible to the antiinflammatory effect of peripheral organs when the CAP is activated and fails to reflect the regional cardiac inflammation [36,37]. Thus, immunohistochemistry was performed to examine cardiac macrophage infiltration. In a prior study, stimulating the CAP was shown to promote the transformation of Ly-6C high macrophages to Ly-6C low macrophages in the late infarction phase [17]. Similarly, our results showed that cholinergic evocation reduced the accumulation of M1 macrophages and increased M2 macrophages infiltration in the peri-infarct area. A reasonable explanation for this result is that anti-inflammatory cytokines released from M2 macrophages limit the activation and proliferation of M1 macrophages [38]. Interestingly, a question originating from our study is why M1 and M2 macrophages maintain the high abundances in the Vag group. This effect may occur because the biological functions of recruited macrophages at injured sites as proinflammation or repair are dependent on microenvironmental stress [39]; that is, the enhanced macrophage plasticity synchronously tends to the proinflammation and anti-inflammation preponderances in the absence of vagus [40]. In terms of the pathological results, upregulating the CAP ameliorated cardiac fibrosis in the peri-infarction area, while vagus transection partially offsets this protective effect. Previous evidence supported that macrophage polarization plays an important role in the process of cardiac fibrosis [41]. In our study, the decrease in M1 macrophages induced by cholinergic elicitation may mediate attenuation of extracellular matrix remodeling by downregulating the expression of matrix metalloproteinase, TNF-α, and IL-1 [42].
Most importantly, we reported the ameliorative effect of CAP activation on mitochondrial injuries under CHF conditions for the first time. In the context of chronic inflammation, detrimental cytokines such as TNF-α trigger mitochondrial dysfunction via inducing ROS overproduction and further beget a series of downstream events as apoptosis and hyperinflammation [8]. Mitochondria-mediated apoptosis relies on the inhibition of Bcl-2 and the activation  Mediators of Inflammation of Bax and caspase-3/7 [43]. Here, we observed a decline in ROS overload in the PNU group and the alleviation of cardiomyocyte apoptosis, especially the mitochondria-mediated apoptosis, according to changes in mitochondrial apoptosisrelated proteins such as caspase-3, Bcl-2, and Bax. Different periods of ongoing mitophagy were discovered by TEM, including the segregation of damaged mitochondria and degradation mediated by lysosomes. Regrettably, no further investigation was conducted in allusion to mitophagy in the present study. It has been widely reported that steady mitochondrial dynamics confer extensive advantages to mitochondrial biogenesis and cellular events [44]. Thereupon, we detected the changes in key proteins implicated in mitochondrial dynamics under CAP initiation. Confusingly, the expression of both Mfn1 and Opa1 was upregulated in the PNU group and downregulated in the Vag group, while Fis1 but not Drp1 expression changed after cholinergic interventions. We merely speculated that CAP induction potentially drove more healthy mitochondrial fragments toward rebirth through mitochondrial fusion mediated by Mfn1 and Opa1 [45], but the exact mechanisms remain unclear.
In consideration of the beneficial effect of CAP elicitation, numerous studies have focused on the activated JAK/ STAT3 pathway and NF-κB/HMGB1/RAGE pathway in immune cells [46]. Our study suggested that the activation of STAT3 and inactivation of NF-κB were implicated in the cardioprotective effects of CAP stimulation, possibly due to the negative modulations of these two transcription factors on downstream phenotypes like inflammation, apoptosis, and MPTP activities, through the signaling cascade triggered by TNF-α/TNFR [47].
Finally, an in vitro cell experiment was performed to validate the association of mitochondrial injury with proinflammatory cytokines. As expected, together with the reductions in MMP and mPTP opening, the major event of mitochondrial apoptosis [48] was increased markedly when   Mediators of Inflammation cells were exposed to the exogenous TNF-α. These results are consistent with other findings, as evidenced by the enhanced mitochondrial apoptosis in the CHF and Vag groups in vivo. Consistently, the expression of mitochondrial fusion-related proteins as Mfn1 and Opa1 was downregulated when challenged by TNF-α. However, we observed that TNF-α had no effect on cell proliferations. The present in vitro experiment may imitate a systemic scenario in which hyperinflammation contributes to mitochondrial dysfunction and accelerates a vicious pathophysiological cycle.

Limitations
There were a few limitations in our study. First, to highlight the protective effect of cholinergic elicitation at any one time point, we ignored the influences of time factor on several repeated measurement data and used univariate ANOVA to analyze effects of indicators such as metabolisms and cytokines. Second, the method of CAP modulation was based on the methods of previously published work, and it was still unclear whether the different degrees of activating or inhibiting the CAP had different effects on the expected outcome. Third, the results of the cell experiment only explained the detrimental impact of exogenous proinflammatory cytokine on mitochondria and indirectly indicated that eliciting the CAP prevented mitochondrial injury linked to inflammation. However, a better design should be aimed at investigating the protective effect of CAP induction on HF cell models. Fourth, our results merely affirmed the protective role of CAP stimulation on CHF induced by MI, but the therapeutic effects on cardiac remodeling caused by other pathogenic factors, such as hypertrophy and cardiotoxicity, remain unknown. Fifth, our results were inclined to the onefold observations of efficacy produced by eliciting CAP and the descriptions of altered pathophysiological phenomena, whereas the key target of cholinergic stimulation in ameliorating mitochondrial dysfunction and the exact molecular mechanisms were still not elucidated.

13
Mediators of Inflammation 14 Mediators of Inflammation [53][54][55]. Spurred by our promising results, further studies should focus more on the clinical translation of cholinergic elicitation in CHF populations.

Conclusion
This study provides evidence that cholinergic elicitation exerts a salutary effect by suppressing the production of proinflammatory cytokines, attenuating cardiac fibrosis, and potentially ameliorating inflammation-linked myocardial mitochondrial injuries. These alterations above are associated with the holistic inflammation resolution, systemic energy metabolism balance, and improvements in cardiac dysfunction.

Data Availability
The datasets of this study are available from the corresponding authors on demand.