The mitochondria-associated membrane protein PACS2 promotes mitophagy and energy metabolism to maintain right cardiac function in hypobaric hypoxia exposure

Hypobaric hypoxia (HH) is the primary challenge at high altitude. Prolonged HH exposure impairs right cardiac function. Mitochondria-associated membrane (MAM) plays a principal role in regulating mitochondrial function under hypoxic conditions, but the mechanism remains poorly understood. In this study, proteomics analysis identi�ed that PACS2, a key protein in MAM, and mitophagy were down-regulated in HH conditions. Metabolomics analysis indicated suppression of aerobic oxidation of glucose and fatty acids. Cardiomyocyte Pacs2 de�ciency disrupted MAM formation and endoplasmic reticulum (ER)-mitochondria calcium �ux further inhibiting mitophagy and mitochondrial energy metabolism during HH exposure. Overexpression of Pacs2 reversed these effects. Cardiac-specic knockout of Pacs2 exacerbated mitophagy inhibition, cardiomyocyte injury and right cardiac dysfunction induced by HH. Knock-in of Pacs2 recovered HH-induced RV structural and functional impairment. Thus, PACS2 is essential for protecting cardiomyocytes through mechanisms of ER-mitochondria calcium �ux, mitophagy, and mitochondrial energy metabolism, thereby maintaining right cardiac function at high altitude.


Introduction
High-altitude areas cover much of the total geographical area worldwide, and human exposure to high altitudes is increasing for various reasons.Hypobaric hypoxia (HH) caused by increasing altitude is the main physiological challenge in such conditions and has long been recognised as a cause of cardiac stress.Acute exposure to high altitudes induces an increase in the right ventricular (RV) afterload, leading to alteration of the RV lling patterns 1 .Prolonged exposure to high altitude conditions further results in chronic remodelling of the cardiac structure and function, ultimately leading to right heart failure 2,3 .Among these cardiac adaptive and/or pathological alterations, cardiomyocyte responses, particularly in intracellular homeostasis maintenance during hypoxia, are the critical molecular basis that determine the adaptive cardiac outcomes 4 .
Cardiomyocytes consume majority of oxygen in the mitochondria as an electron donor for oxidative phosphorylation (OXPHOS) 5 .Thus, mitochondria are highly sensitive to decreases in oxygen levels in cardiomyocytes.Hypoxia increases oxidative stress and mitochondrial DNA mutation and causes mitochondrial dysfunction 6 .Hypoxia also suppresses mitochondrial OXPHOS and leads to the accumulation of aerobic metabolic substrates and anaerobic metabolites, which result in cardiomyocyte injury and cardiac dysfunction 7 .These damaged mitochondria and excessive metabolic substrates could be removed by mitophagy 8,9 .However, mitophagy is suppressed in some cardiac diseases, such as diabetic cardiomyopathy and ischemic cardiomyopathy 10 , which leads to mitochondrial dyshomeostasis and cardiac dysfunction.Therefore, an appropriate level of mitophagy serves as a protective mechanism to maintain the mitochondrial function in response to cardiac stress 11 .In recent years, several mitochondrial membrane receptors containing the LC3-interacting region (LIR) motif have been found to mediate mitophagosome formation under acute hypoxic conditions 8, 12 .However, there is a paucity of information regarding cardiomyocyte mitophagy during chronic HH exposure and the precise mechanism has not been fully elucidated.
Recent studies indicated that cardiomyocyte mitochondrial function is regulated by the mitochondriaassociated membrane (MAM) structure, tethering two organelles by proteins located on opposing membranes 13,14 .Stable contact between the endoplasmic reticulum (ER) and the mitochondria integrates the two organelles' functions to regulate autophagy 15 , calcium signalling 16 , and others.Phosphofurin acidic cluster sorting protein 2 (PACS2) is a key physical linkage protein on the MAM, connecting the ER with the mitochondria and maintaining MAM stability.The physiological function of the MAM, particularly the calcium ux from the ER to the mitochondria and the transport of membrane phospholipids, relies on PACS2 17 .Previous studies have reported the important role of PACS2 in the development and progression of several tumours [18][19][20] .Tumour cells survive in a hypoxic microenvironment because of their in nite proliferation.Therefore, PACS2 may be an effector of hypoxia and may also be involved in the cell responses at high-altitude conditions, characterised by HH.
In this study, we established HH conditions to closely simulate high-altitude exposure and focused on the PACS2-mediated mitophagy and mitochondrial energy metabolism, regarding the calcium ux across the MAM as the core mechanism.This study provided insight into the cardiomyocyte response to HH conditions.We also interpreted the mechanism underlying high-altitude-induced right cardiac dysfunction.

Results
Proteomics revealed down-regulated PACS2, mitophagy, and mitochondrial energy metabolism in the right myocardium in response to hypobaric hypoxia.C57BL/6J mice were assigned to a HH chamber to simulate high-altitude conditions for 6 weeks (Fig. 1a).The right cardiac function of the mice was detected by echocardiography and RHC.As depicted in Figure 2, the mice subjected to HH conditions showed a markedly lower RV FAC (Fig. 2a, d) and VTI (Fig. 2h), concurrent with an increase in the mPAP (Fig. 2c, f), Tei index (Fig. 2b, e), and max dP/dt (Fig. 2g), indicating impaired right cardiac function.
Additionally, increased heart mass (Fig. 2i, j) and Fulton index (Fig. 2k) were observed after the 6-week HH exposure.The abovementioned results indicated a stable RV dysfunction generated after 6-week chronic HH exposure.
Next, we obtained the right myocardium and performed proteomics and metabolomics analyses, accordingly.The different candidates were de ned using a criterion of ≥ 1.2-fold change and a signi cant difference between the groups.In the proteomics analysis, we identi ed 217 down-regulated proteins and 82 up-regulated proteins in the right myocardium of HH-exposed mice compared with the respective levels in the NN counterparts (Fig. 1b, Supplemental Table 1).In the metabolomics analysis, we identi ed that 53 endogenous metabolites increased and 22 decreased (Supplemental Table 2).Hierarchical clustering analysis of metabolomics indicated markedly altered cardiac metabolic pathways under HH exposure (Fig. 1c).Among the differential expressed proteins, we identi ed PACS2 as being signi cantly down-regulated by 3.16 folds (Fig. 1d).In addition to PACS2, MAP1LC3A, MAP1LC3B, autophagy-related 16-like 1, and sequestosome 1 (SQSTM1/p62) were remarkably down-regulated, which were involved in phagophore formation and mitophagy induction (Fig. 1e).To gain insight into the possible biological impact of HH exposure, we subjected proteins that were down-or upregulated to KEGG and GO pathway enrichment analysis.The glycolysis, HIF-1 signaling pathway, and focal adhesion were up-regulated during HH, while the OXPHOS, citrate cycle, fatty acid beta-oxidation, and mitophagy were down-regulated (Fig. 1f and 1g).The abovementioned results indicated that both mitophagy and mitochondrial energy metabolism were impaired in the right myocardium of the mice due to HH exposure.Cardiac Pacs2 ablation exacerbated the right cardiac dysfunction and structure impairment induced by hypobaric hypoxia exposure.To determine the role of PACS2 in maintaining right cardiac function and structure, we generated cardiomyocyte-speci c Pacs2 −/− (Pacs2 ox/ ox /Cre αMHC+/− ) mice.The Pacs2 −/− mice and their littermate controls (Pacs2 wt/+ /Cre αMHC+/− ) underwent echocardiography and RHC evaluation after a 6-week HH exposure.During exposure to HH, Pacs2 −/− mice exhibited a signi cantly lower RV FAC (Fig. 2a, d) and higher Tei index (Fig. 2b, e) than their littermate controls.Interestingly, we observed that other parameters re ecting the afterload, including the mPAP (Fig. 2c, f), max dP/dt (Fig. 2g), and RV VTI (Fig. 2h), were not further aggravated in Pacs2 −/− mice during HH exposure.The heart weight was not further changed in Pacs2 −/− mice (Fig. 2j); however, cardiomyocyte-speci c knocking out of PACS2 resulted in right ventricular hypertrophy based on a signi cant increase in the Fulton Index (Fig. 2k).HE staining further revealed increased RV chamber thickness, decreased RV chamber size, and disordered arrangement of RV myocardium caused by cardiac Pacs2 ablation (Fig. 3a).Masson's trichrome staining demonstrated signi cant collagen deposition in the right myocardial interstitial space after HH exposure (Fig. 3b, c).The mean cross-sectional area (CSA) of the RV cardiomyocytes in the HH group was signi cantly larger than that in the NN group (Fig. 3d, e).Such substantial right cardiac remodelling caused by HH was signi cantly more serious in the hearts of Pacs2 −/− mice.The cardiomyocyte injury was also evidenced by plasma markers (BNP, TnI, and CK-MB), which were much higher in Pacs2 −/− mice than in the remaining two groups (Fig. 3f-h).Our results indicated that HH-induced right cardiac impairment phenotype became more noticeable following Pacs2 ablation.
To verify the separate role of Pacs2 ablation, we evaluated the right cardiac function under NN conditions.Compared to littermate controls, Pacs2 −/− mice showed a normal mPAP (Supplemental Fig. 1c, 1f-h), which was accompanied by impaired right cardiac function, as revealed by lower RV FAC (Supplemental Fig. 1a, d) and an increased Tei index (Supplemental Fig. 1b, e).Pacs2 −/− mice in NN exhibited signi cantly increased myocardial disorder and brosis (Supplemental Fig. 2a-c).Additionally, cardiomyocyte injury was also evident in NN conditions in the presence of Pacs2 −/− (Supplemental Fig. 2d-h).In general, Pacs2 ablation exacerbated the cardiomyocyte injury and right cardiac dysfunction; however, it did not act on the RV afterload during HH exposure.hypoxia.To determine how PACS2 responds to HH, we compared the subcellular localization of MAMassociated proteins in isolated right myocardium under HH or NN conditions.As shown in Fig. 4a, different fractions were identi ed with the following organelle markers: FACL4, VDAC1, MFN2, FIS1, TOMM20, CNX, and DRP1.The level of the PACS2 in MAM signi cantly decreased in the HH group compared with its levels in the NN counterparts, although a small amount of PACS2 can also be found in the cytosol.However, the levels of other MAM-related proteins were not noticeably altered in MAM fractions isolated from the HH group.To assess whether PACS2 affects MAM integrity, we examined the ER-mitochondrial contacts in Pacs2 −/− myocardium.As illustrated by the TEM images (Fig. 4b, c), the proportion of ER in close contact with mitochondria relative to the total ER content was lower in the HH group than in the NN group and further decreased in the Pacs2 −/− mice.Consistent with the TEM images, immuno uorescence analysis clearly showed a lower level of co-localization of the ER with mitochondria in Pacs2 −/− mice than in the remaining groups (Fig. 4d, e).
Next, we determined whether PACS2 alteration affected mitophagy.Impaired mitophagy induced by HH exposure was con rmed by decreased mitophagy markers in the right myocardium.The blotting results indicated that the level of MAP1LC3B-II was lower in Pacs2 −/− mice than in the controls during HH conditions (Fig. 4f, g).Moreover, immunostaining analysis revealed that Pacs2 deletion further reduced the co-localization of MAP1LC3B puncta and mitochondria induced by HH exposure (Fig. 4h, i).Taken together, cardiac Pacs2 ablation exacerbated MAM disruption and mitophagy reduction induced by HH.
Hypobaric hypoxia reduced MAM formation and mitophagy in vitro.To explore the impact of HH on the MAM structure and biological function, we also measured the levels of MAM-related proteins in H9C2 cardiomyocytes exposed to simulated hypobaric hypoxia in vitro.As depicted in Fig. 5a, in line with the in vivo results above, the expression of PACS2 in the MAM from HH-treated cells was lower than that in the MAM from NN-treated cells.Consistently, confocal imaging and Pearson's correlation coe cient analysis (Fig. 5b, c) showed a decreased association between the ER and mitochondria in simulated HH-treated cardiomyocytes compared with that in the NN cells.TEM imaging showed swollen mitochondria and fewer mitochondria adjacent to the ER after HH exposure (Fig. 5d, e).These data suggested that HH decreases the MAM junction structure in cardiomyocytes.Furthermore, we evaluated mitophagy levels in H9C2 cardiomyocytes.We found decreased MAP1LC3B-II transfer (Fig. 5f, g) and co-localization with mitochondria after HH exposure (Fig. 5h, i).To further verify the impaired mitophagy, we transfected pH-dependent mitochondrial protein Keima into the cardiomyocytes, which can shift from green to red as mitochondria are delivered to lysosomes.LSCM monitoring showed that HH induced a markedly decreased mitophagy index in the cardiomyocytes (Fig. 5j, k), indicating that HH decreased the number of mitophagosomes and impaired the mitophagy ux.The above results in the H9C2 cell lines as well show that HH reduces MAM formation and mitophagy.
Hypobaric hypoxia reduced ER-mitochondria calcium ux and mitochondrial oxidative phosphorylation in vitro.PACS2 was reported to maintain the junction of the MAM and regulate mitochondrial calcium ux 21 .In this study, we found that [Ca 2+ ] m in the H9C2 cardiomyocytes after HH treatment was markedly lower than that in the NN group (Supplemental Fig. 3a, b).To determine the origin of mitochondrial calcium, we incubated H9C2 cardiomyocytes under HH or NN with a cytoplasmic Ca 2+ chelator, BAPTA-AM (10 µM), in calcium-free HBSS for 10 min.Mitochondrial calcium was labelled by a Rhod2-AM probe, and the cells were observed and measured under LSCM.TG-(an inhibitor of calcium pump, Fig. 6a, b) and ATP-(an indirect IP3R agonist, Fig. 6c, d) elicited ER-mitochondria calcium ux was lower in the HHtreated cells than in NN-treated cells.Inositol trisphosphate receptors (IP3R) are important ER calcium release channels 22 .Therefore, we added 2-APB, which blocked the release of calcium from IP3R.Comparable mitochondrial calcium to the HH exposure was observed when treated with 2-APB (Fig. 6e, f), suggesting that IP3R is required for maintaining the physiological mitochondrial calcium levels under NN conditions.
MAM formation and ER-mitochondrial calcium ux are essential for mitochondrial energy metabolism.Thus, we evaluated the effects of HH on cardiomyocyte mitochondrial energy metabolism using a Seahorse XF analyser to measure the mitochondrial respiration and glycolytic ux.We found that cardiomyocytes exhibited signi cant decreases in basal and maximal cellular oxygen consumption rate (OCR) in response to HH. ATP production and spare respiration capacity were also signi cantly lower after HH exposure (Fig. 6g, h).The extracellular acidi cation rate (ECAR) results indicated an increase in glycolysis and glycolytic capacity due to insu cient oxygen (Fig. 6i, j).In addition, OCR measurement with a medium containing BSA-conjugated palmitic acid signi cantly decreased after HH exposure (Fig. 6k, l).The abovementioned results demonstrate that after HH exposure, the cardiomyocytes displayed a metabolic reprogramming, which is represented by the restriction of FAO-related OXPHOS and a tendency to rely more on glycolysis than aerobic glycolysis for adapting to the HH condition.Considering that HH caused a decline in ER-mitochondria calcium ux, one mechanism that potentially accounts for the metabolic shift may be associated with the regulation of mitochondrial calcium.ER-mitochondria calcium ux is involved in PACS2-mediated mitophagy and mitochondrial energy metabolism.We next determined the contributions of ER-mitochondria calcium in PACS2-induced mitophagy and mitochondrial energy metabolism.We obtained cardiomyocytes with stable overexpression of Pacs2 by lentiviral vector (LVV) infection (Fig. 7a).ER-mitochondria contacts increased in cells where Pacs2 was overexpressed (Fig. 7b, c).We found reversed [Ca 2+ ] m in LVVs-infected cardiomyocytes (Supplemental Fig. 3a, b).Similar results were observed in cultured cells with Pacs2 overexpression in the dynamics of mitochondrial calcium ux (Fig. 7d, e), indicating a source of calcium ux released from the ER.The regulation of calcium ux between the ER and mitochondria via IP3R is a major function of the MAM 23 .As depicted in Fig. 7f and 7g, the restored calcium levels caused by the overexpression of Pacs2 were partly blocked by 2-APB, suggesting that Pacs2 overexpression promoted ER calcium release in the MAM through IP3R.
Additionally, higher MAP1LC3B-II levels were observed after LVVs-overexpression of Pacs2 (Fig. 7h), which could also be blocked by 2-APB (Fig. 7i), suggesting that the PACS2mediated ERmitochondria calcium ux was required for mitophagy.With the supplementation of PACS2, more MAP1LC3B puncta co-localized with mitochondria (Fig. 8a, b) and an increased mitophagy index were observed in HH conditions (Fig. 8c, d).These data demonstrated that PACS2 restored impaired mitophagy through enhanced ER-mitochondria calcium ux.To investigate whether ERmitochondria calcium ux was also involved in PACS2-mediated mitochondrial energy metabolism alteration, we compared the real-time changes in OCR and ECAR in the H9C2 cardiomyocytes with or without overexpression of PACS2 under HH treatment.With PACS2 supplementation, the declined basal respiration, ATP production, and the maximal respiration (Fig. 8e, f) as well as increased basal and maximal ECAR (Fig. 8g, h) induced by HH were signi cantly reversed.The recovery in mitochondrial respiration was also blocked by 2-APB treatment (Fig. 8e, g).To extend the hypothesis that PACS2 recovered OCR, which is supported by FAO, we further measured OCR in a medium containing palmitate-BSA as an exogenous FAO substrate.Notably, the cardiomyocytes showed a reversed OCR after the supplementation of PACS2 when compared with an empty vector control (Fig. 8i, j), which was signi cantly blocked on adding 2-APB.This indicates that PACS2 enabled HH-treated cardiomyocytes to switch from glycolysis to an increased reliance on FAO for ATP production, and this metabolic reprogramming at least partly depends on the calcium ux across MAM.Taken together, these data suggested that ER-mitochondria calcium ux was essential for PACS2mediated mitophagy maintenance and mitochondrial energy metabolism after HH exposure.Cardiac Pacs2 knock-in alleviated HH-induced right cardiac dysfunction.LVVs-overexpression of Pacs2 signi cantly reversed MAM formation, mitophagy, and mitochondrial energy metabolism in cardiomyocytes in vitro.To verify the contributions of PACS2 in maintaining RV function during HH exposure in vivo, we generated cardiomyocyte-speci c Pacs2 knock-in mice.Histological analysis of the hearts from the Pacs2 knock-in mice showed signi cantly decreased right cardiac hypertrophy (Fig. 8a), cardiac brosis area (Fig. 8b, c), and cardiomyocytes CSA (Fig. 8d, e) with HH exposure.In addition, the Pacs2 knock-in mice had lower plasma levels of BNP, TnI, and CK-MB (Fig. 8f-h) than that of their littermate controls, indicating that PACS2 supplementation reduced the HH-induced myocardial damage.Compared with their littermate controls, the Pacs2 knock-in mice exhibited a higher RV FAC (Fig. 9a, d) and lower Tei index (Fig. 9b, e).As expected, Pacs2 overexpression failed to alter mPAP (Fig. 9c, f), max dP/dt (Fig. 9g), and RV VTI (Fig. 9h) during HH exposure.These data suggested that conditional Pacs2 knock-in reduced cardiomyocyte injury and partially recovered RV cardiac function after HH exposure without signi cantly in uencing the RV afterload.

Discussion
This is the rst study to reveal the underlying mechanism of PACS2 in HH-mediated cardiomyocyte injury and right cardiac dysfunction.The core process was the down-regulated PACS2 localized in MAM after HH exposure.PACS2 reduction further suppressed MAM formation and resulted in decreased calcium ux from the ER to the mitochondria via the IP3R calcium channel.The reduction of mitochondrial calcium in ux further inhibited mitophagy and mitochondrial energy metabolism, inducing cardiomyocyte injury and right cardiac dysfunction.Moreover, cardiomyocyte-speci c knock-in of Pacs2 reversed right cardiac dysfunction and RV brosis.Of note, neither knockout nor knock-in of Pacs2 in uenced the RV afterload, highlighting an independent role of PACS2-directed cardiomyocyte responses in maintaining right cardiac function.Thus, our results provided potential therapeutic targets for high-altitude induced right cardiac impairment.
Su cient oxygen supply is the most essential condition for the survival and function of cardiomyocytes.Hypoxia induces pulmonary vasoconstriction and increases pulmonary vascular resistance.Prolonged hypoxia further results in right cardiac function impairment and even right heart failure 24 .Thus, in this study, we focused on the right rather than on the left cardiac function.Hypoxia is usually generated by normobaric hypoxia (NH) or HH in the experiments.NH lowers the partial pressure of inspired oxygen (PiO 2 ) by reducing the fraction of inspired oxygen through the addition of exogenous nitrogen without altering the barometric pressure.Conversely, HH lowers the PiO 2 by reducing the barometric pressure.
Previous studies have suggested that NH and HH induced similar cardiac adaptations over a short duration, although lower SpO 2 and worse right cardiac function emerged during long-term exposure 25,26 .
Thus, more complicated mechanisms may exist in HH than in NH, including intravascular bubble formation, increased alveolar deadspace, altered uid permeability, and a mismatch in ventilation and perfusion 25,27 .Our study established a long-term HH exposure model to simulate real cardiac function alteration and cardiomyocyte response in high-altitude environments.Previous studies on cardiomyocyte injury caused by hypoxia mainly focused on the increase in oxygen free radicals and anaerobic metabolites, eventually leading to cardiomyocyte apoptosis, myocardial brosis, and irreversible cardiac remodelling 28,29 .Our study rstly indicated that both mitophagy and mitochondrial energy metabolism were involved in cardiomyocytes survival under HH conditions.Our results provided a novel mechanism that results in right cardiac dysfunction at high altitudes.
Previous studies reported that PACS2 is closely associated with the onset and progression of tumours, such as colorectal tumour and liver cancer 18,30 .Due to their in nite proliferation ability, tumour cells survive in relative hypoxic conditions.Therefore, it is essential to understand the biological function of PACS2 in regulating cell fate in response to hypoxic conditions.Accordingly, after HH exposure, PACS2 expression was found to be markedly reduced in the MAM, although with a moderate increase in PACS2 in the cytosol both in vivo and in vitro (Fig. 4a and Fig. 5a).This implied a dynamics translocation from the MAM to the cytosol, which may partly explain the decreased PACS2 in the MAM.The free form of PACS2 in the cytosol contributed to nuclear gene expression and membrane tra cking rather than calcium ux, thus exhibiting the suppression of downstream mitophagy and energy metabolism.Besides PACS2, recent studies have found that the FUN14 domain containing 1 (FUNDC1), a new protein in the MAM, is responsible for the release of calcium from the ER to the mitochondria and mitophagy induction in mouse cardiomyocytes 31,32 .Notably, our results revealed that PACS2 served as a new mitophagic regulator in the MAM and also modulated the release of calcium from ER to mitochondria in cardiomyocytes.However, the precise molecular mechanism of how PACS2 cooperates with FUNDC1 to regulate calcium ux, mitophagy, and cardiac function remains unknown.We considered that a few special proteins in the MAM may interact to form a protein complex or "protein machine" and get involved in the abovementioned process.PACS2 may be the key protein and serves as a scaffold to sponge other proteins.
Mitophagy is essential for mitochondrial homeostasis and quality control in cardiomyocytes 33 .During hypoxia, mitophagy is the sole mechanism by which cardiomyocytes eliminate super uous or damaged mitochondria 11 .However, the mechanisms underlying mitophagy remain largely unknown.Previous studies on mitophagy have focused on several protein receptors, including BCL2 interacting protein 3 (BNIP3), BNIP3-like, and FUNDC1, on the mitochondrial membrane.Most of them have a classic LIR motif to directly bind MAP1LC3B for mitophagy activation 34 .In this study, we found a new protein in the MAM without this classic LIR motif, although it was closely associated with HH-mediated mitophagy.PACS2 did not directly link to autophagy-associated proteins, such as ATG5, ATG7, Beclin1, and MAP1LC3B; however, it acted as a calcium channel to promote calcium in ux into the mitochondria 21 .Usually, intracellular calcium is considered an activator of autophagy 35,36 .
To date, the role of calcium signaling in autophagy regulation is highly controversial.Most studies considered that calcium works as an activator of autophagy because calcium mobilizing agents and calcium ionophores promote autophagy by elevating intracellular calcium concentration 37 .In this study, we detected that PACS2-mediated calcium in ux was required for HH-induced mitophagy in cardiomyocytes, which further veri ed the impact of intracellular calcium on mitophagy regulation.However, the way in which mitochondrial calcium is involved in mitophagy activation requires further exploration.Mitochondrial calcium uptake occurs mostly through MAM, which closely contacts with the ER and renders a micro-domain with a su ciently high calcium concentration 38 .A recent study reported that mitochondrial calcium in ux inhibition decreased ATP production, enhanced mitophagy, and provided cardioprotection in cardiac failure 39 .Conversely, we found that HH decreased ER-mitochondria calcium ux and calcium-mediated mitophagy.Similar to our results, Böckler and Zou demonstrated that ERmitochondria contacts and calcium ux across the MAM were required for autophagic removal of mitochondria, since arti cially tethering ER and mitochondria rescued mitophagy defects 16,31 .Besides mediating calcium ux, the ER-mitochondria encounter structure may also supply the growing phagophore with lipids synthesised in the ER, which then enclose the impaired mitochondria to form a mitophagosome.Hence, ER-mitochondria-mediated calcium ux is required for mitophagy induction.
In addition to the abovementioned role of calcium in mitophagy, mounting evidence suggests that calcium also dynamically regulates the aerobic energy metabolism by stimulating mitochondrial OXPHOS 40,41 .In highly energy-consuming tissues, such as the heart, OXPHOS in the mitochondria provides a major source of cellular ATP through the oxidation of substrates, including fatty acids, glucose, and ketones 42 .We found that to adapt to the HH condition, cardiomyocytes mainly rely on the glycolytic pathway rather than the OXPHOS pathway.As a critical signaling molecule in mitochondrial energy conversion, su cient mitochondrial calcium concentration is required to activate the mitochondrial dehydrogenases including the pyruvate dehydrogenase complex (PDHC), NADH-isocitrate dehydrogenase (ICDH) and α-ketoglutarate dehydrogenase (α-KGDH) 43,44 .Other components within the energy-producing pathways besides NADH generation, such as downstream ATPase and the cytochrome chain, were also signi cantly stimulated by calcium 45,46 .Current studies supported the proposed physiological metabolic role for calcium entry into mitochondria matrix through mitochondrial calcium uniporter (MCU) complex 47 .Together with some recent reports, we indicated that the IP3R channels were also associated with alterations in mitochondrial calcium ux, especially in cardiomyocytes under HH exposure.IP3R-mediated calcium signaling required the quanti cation of PACS2 and the proximity of the ER and mitochondria.The supplement of PACS2 could improve mitochondrial respiration e ciency during HH exposure.Our previous randomized double-blinded clinical trial proposed that cardiac function could be recovered by optimizing myocardial energy metabolism 48 .Combined with those results, our present results provide a therapy for improving cardiac function at high altitude targeted on the energy metabolic reprogramming based on calcium ux across MAM in cardiomyocytes.Moreover, linking the reprogramming of energy metabolism induced by PACS2 supplement was associated with enhanced mitophagy.We considered that mitophagy may, at least partly, provide relatively e cient substrates such as fatty acids for maintaining energy demand.However, the exact role of calcium in cardiomyocyte energy metabolism reprogramming requires con rmation using accurate techniques such as isotope tracing analysis.
In conclusion, we described a novel cardiomyocyte injury mechanism during HH exposure to simulate high altitude.HH down-regulated the expression of PACS2 in MAM.Decreased PACS2 disrupted MAM formation and calcium transfer from the ER to the mitochondria, leading to mitophagy inhibition and mitochondrial energy metabolism impairment, which induced cardiomyocyte injury and right cardiac dysfunction during HH exposure.Our study identi ed a potential target for the prevention and treatment of cardiovascular diseases caused by high-altitude exposure.

Methods
Development of chronic hypobaric hypoxia-induced mouse model.All animal procedures were approved by the Experimental Animal Ethics Committee of the Army Medical University and conformed to the regulations of the Guide for the Care and Use of Laboratory Animals.Male C57BL/6J mice (6-8 weeks old) were housed in a temperature-controlled environment with a 12-hour light/dark cycle.The mice were subjected to HH condition (10.0%oxygen content and 46.3 kPa air pressure) in a chamber (AIPUINS XBS-03, Hangzhou, China) or were housed in normobaric normoxia (NN, 20.9% oxygen content and 101.3 kPa air pressure) as controls for 6 weeks (Fig. 1a).
Blood preparation and enzyme-linked immunosorbent assay.Detection kits for mouse plasma B-type natriuretic peptide (BNP), troponin I (TnI), and creatine kinase-MB (CK-MB) were purchased from Jiangsu Jingmei Biological Technology Co., Ltd.(Jiangsu, China).Approximately 1.5 mL of blood was drawn from each mouse and stored in procoagulant tubes.Plasma was separated by centrifugation (3000×g, 20 min) after coagulation at room temperature for 10 min.The plasma levels of BNP, TnI, and CK-MB were measured using a commercially available BNP enzyme-linked immunosorbent assay (ELISA) kit (JM-02343M2, 210727B8), TnI ELISA kit (JM-02662M2, 210727I4), and CK-MB ELISA kit (JM-03084M2, 210727C6), respectively, following the manufacturer's instructions.Hemodynamic monitoring.Right heart catheterization (RHC) was performed using a pressure detecting device (ADInstruments Mikro-Tip®, MPVS Ultra RSBMIL002/M) after a 6-week HH or NN exposure.The mice were placed on a heated pad and anesthetized with 2% iso urane.The right jugular vein was exposed, and a 1F needle (ADInstruments Mikro-Tip®, SPR-1000) was slightly bent inwards to conduct the cannula containing the catheter into the jugular vein.The cannula was maneuvered to the right ventricle, with its tip pointing towards the heart until an RV pressure curve could be identi ed using LabChart 7 software.Next, the cannula tip was manipulated to the left and upward.The catheter was advanced into the main pulmonary artery, passing through the pulmonary valve.When the catheter enters the main pulmonary artery, the diastolic pressure rises on the monitor, and a pulmonary artery pressure curve appears.When the curve was constant, the related indices, such as the mean pulmonary artery pressure (mPAP), maximum positive time derivative of left ventricular pressure (max dP/dt), and RV velocity time integral (VTI) and electrocardiograms were measured.
Evaluation of right ventricular hypertrophy.After the hemodynamic measurement, the mice were sacri ced by cervical dislocation, and their hearts were removed quickly and weighed.The free wall of the right ventricle was dissected from the left ventricle and interstitial septum.Whole heart weight (normalized by body weight) and Fulton's index (right ventricle / [left ventricle + interstitial septum]) were used as indices of cardiac hypertrophy.Histological analysis.The hearts from the mice exposed to NN and HH were excised, placed in 4% paraformaldehyde, dehydrated in graded concentrations of ethanol, immersed in xylene, and embedded in para n.Sections of 5-µm thickness were cut on a microtome with a disposable blade, stained with hematoxylin-eosin and Masson's trichrome stain, and examined by light microscopy.The cardiomyocyte cross-sectional area (CSA) was analyzed by staining the heart sections with a wheat germ agglutinin-Alexa Fluor® 647 conjugate (W32466, Invitrogen).Six mice from each group were included in the histological analysis.A minimum of ve cross-sections of each heart were examined, and the measurements were averaged for statistical analysis.ImageJ software (RRID:SCR_003070) was used to quantify all the histological endpoints.
Generation of cardiomyocyte-speci cPacs2knockout mice.Cardiomyocyte-speci c Pacs2 knockout (Pacs2 −/− ) mice were generated on a C57BL/6J background by the CRISPR/Cas9 system at cyagen.The gRNA to mouse Pacs2 gene, the donor vector containing loxP sites, and Cas9 mRNA were co-injected into fertilized mouse eggs to generate targeted conditional knockout offspring.Pacs2 ox/ ox mice in which the Pacs2 gene was anked by loxP sites within introns 1 and 3 (KO region: ~1842 bp) were crossed with αmyosin heavy chain (α-MHC) promoter-Cre transgenic mice (Cyagen Biosciences) to obtain Pacs2 ox/+ /Cre αMHC+/− mice.F0 founder animals were identi ed by PCR followed by sequence analysis, which were bred to wildtype mice to test germline transmission and F1 animal generation.F1 founders, including Cardiomyocyte-speci c Pacs2 knockout (Pacs2 ox/ ox /Cre αMHC+/− ) mice, were genotyped by tail genomic PCR.Generation of cardiomyocyte-speci cPacs2knock-in mice.The cardiomyocyte-speci c Pacs2 knock-in in C57BL/6J mice was created using CRISPR/Cas-mediated genome engineering (Cyagen Biosciences).The Hipp11 locus is located within an intergenic region between the Eif4enif1 and Drg1 genes on mouse chromosome 11.The mouse Pacs2 gene (NCBI Reference Sequence: NM_001291444.1) is located on mouse chromosome 12.For the KI model, the "alphaMHC_long promoter-Kozak-Mouse Pacs2 CDS-rBG pA" cassette was inserted into Hipp11 locus (~0.7 kb 5' of Eif4enif1 gene and ~4.5 kb 3' of the Drg1 gene).To engineer the targeting vector, homology arms were generated by PCR using BAC clone as the template.Cas9 and gRNA were co-injected into fertilized eggs with a targeting vector for mice production.
The pups were genotyped by PCR followed by sequencing analysis.
Echocardiography.Cardiac geometry and function were examined using ultrasonography (GE Vivid 7 Dimension, L15/6-MHz transducer).The mice were anesthetized with 2% iso urane while maintaining proper body temperature (36-37°C) and heart rate (450-550 beats/ minute).The temporal frame rate in the echo mode was set to 60 Hz.A 1.0-mm sampling gate was used to obtain the in ow and out ow velocities, and the maximal sweep speed was 200 mm/s.RV end-diastolic (ED) and end-systolic (ES) areas were measured using ImageJ from the apical or basal 4-chamber views at end-diastole or endsystole.The RV fractional area change (FAC) was calculated as follows: FAC = ([ED RV area -ES RV area] / ED RV area) × 100%.For Tei index calculation, the tricuspid closure opening time (TCO) and ejection time (ET) were measured from tissue Doppler myocardial velocity images, as follows: Tei index = (TCO -ET) / ET.Data were collected from six mice per group and represented the average of minimum of ve separate scans in a random blind fashion.To avoid bias, the researcher performed all echocardiography procedures blinded to the experimental treatments.
Cell culture and RNA transfection.Rat H9C2 cardiomyocytes (BFN60804388) were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China).Cardiomyocytes were cultivated in Dulbecco's modi ed Eagle's medium (Sigma-Aldrich, Louis, MO, USA) and 10% fetal bovine serum (Hy Clone Laboratories, PA, USA), and supplemented with 1% antibiotic-antimycotic (1000 U/mL penicillin and 100 µg/mL streptomycin).H9C2 cardiomyocytes in the NN group were incubated at 37°C with 5% CO 2 .HH conditions were achieved by using a HH chamber (Billups-Rothenberg) ushed with a preanalyzed gas mixture of 1% O 2 , 5% CO 2 , and 94% N 2 .To maintain cardiomyocyte cultures, the medium was changed every 2 days.LVVs carrying Pacs2 RNA system were constructed by Gene Pharma Technology (Shanghai, China).The LVVs were added to the cells at a multiplicity of infection of 100.The transfection medium was changed 2 days later, and the cells were continuously cultured in fresh medium.
Real-time quantitative reverse transcription-PCR and western blotting were used to detect the e ciency of Pacs2 overexpression in cardiomyocytes.
Measurement of mitochondrial calcium in intact cells.Cardiomyocytes were seeded on glass-bottomed cell culture dishes and incubated with 1 µM of the calcium indicator Rhod2AM (ab142780, Abcam) at 37°C in the dark for 30 min, as per the manufacturer's guidelines.Next, the cells were washed twice with calcium-free HBSS and imaged under a laser scanning confocal microscope (LSCM, Leica TCS-SP5).The uorescence intensity (F) was normalized to the baseline uorescence value F 0 (F/F 0 ) and expressed as mitochondrial calcium concentration ([Ca 2+ ] m ).We measured F max and F min , as previously described 50 .
F max was obtained by perfusion with 10 µM ionomycin and 5 mM CaCl 2 ; F min was measured by perfusion with 10 mM EGTA and 20 µM BAPTA-AM (B1205, Molecular probes) in HBSS.2-APB (ab120124, Abcam), TG (T9033, Sigma-Aldrich), and ATP (A1852, Sigma-Aldrich) were also added to the external solution at a proper nal concentration.The uorescence intensity of Rhod2-AM was measured using LSCM.The uorescence intensity was converted to [Ca 2+ ] using the following formula: [Ca 2+ ] m = K d × (F − F min ) / (F max − F), where K d is the equilibrium dissociation constant of Rhod2 for Ca 2+ , which was 570 nM.
Immuno uorescence.Cardiomyocytes were stained with MitoTracker Deep Red FM (500 nM; M22426, Invitrogen) and xed in 4% paraformaldehyde (P0099, Beyotime Institute of Biotechnology) at room temperature for 10 min.They were then permeabilized with 0.1% Triton 100-X (P0096, Beyotime Institute of Biotechnology) at room temperature for 30 min.Cells were washed with phosphate-buffered saline (PBS) three times and blocked in blocking buffer (P0102, Beyotime Institute of Biotechnology) for immunostaining at 37°C for 30 min.Samples were incubated with anti-MAP1LC3B antibody (1:100) or anti-ERP72 antibody (ab155800, Abcam; 1:100) at 4°C overnight and then washed in PBS twice, before staining with the secondary antibody (31561, Invitrogen; 1:500) at 37°C for 2 h.Co-localization of uorescence was measured at 100-400 Hz under the LSCM.Samples without primary antibodies were used as negative controls.Images were analyzed using LAS X software (Leica) and Image-Pro Plus 5.0 (Media Cybernetics).Co-localization represented in Pearson's correlation coe cient was measured using automatic thresholding, as previously described 51 .
Measurement of mitochondrial bioenergetics and FAO metabolism.The cellular oxygen consumption rate (OCR) and extracellular acidi cation rate (ECAR) were measured using a Seahorse XF96 extracellular ux analyzer (Seahorse Bioscience, North Billerica, USA).Brie y, cells with/without HH exposure or transfected with LVVs-Pacs2 were plated in XF96-well microplates (6000 cells/well, Seahorse Bioscience).After reaching the proper cell density, the cells were incubated with XF assay medium without CO 2 at 37ºC for 1 h.For OCR measurement, cells were then serially exposed to 1 µM oligomycin (mitochondrial/ATP synthase inhibitor), 2 µM of tri uoromethoxy carbonyl cyanide phenylhydrazone (FCCP, a mitochondrial uncoupler), and 0.5 µM of rotenone/antimycin A (respiratory chain inhibitor), provided in the XF Cell Mito Stress Test Kit (Seahorse Bioscience).Three measurements were performed for each cycle (4 min mixing, followed by 3 min detection).Basal respiration, maximal respiration, proton respiration, and coupled respiration were collected using Seahorse XF Extracellular Flux analyzer software following the manufacturer's protocol.To measure the FAO, cells were cultured replaced by FAO assay medium containing palmitate-BSA according to the manufacturer's instructions.Other conditions were consistent with the normal OCR measurement.OCR or ECAR experiments were conducted at 37 ℃ and adjusted pH to 7.4.Following an XF assay, the number of cells were determined and used to normalize OCR and ECAR.
Measurement of mitophagy levels using the mitochondria-targeted Keima reporter.We used the mtKeima reporter to measure the mitophagy levels.Cardiomyocytes were transfected with mitochondria-targeted monomeric Keima-Red-hyg (mtKeima; AM-V0251HM, Medical and Biological Laboratories Co., Ltd.), which contained hygromycin B-resistance gene.Hygromycin B infection was used to screen and obtain cardiomyocytes stably expressing mtKeima; cardiomyocytes were seeded on glass-bottom dishes and observed under an LSCM to evaluate mitophagy levels.The wavelengths of excitation and emission lters used were as follows: cytoplasmic Keima: 488 nm, 650-760 nm, lysosomal Keima: 561 nm, 570-630 nm 52 .Images were analyzed using ImageJ software.Brie y, the cardiomyocytes and mtKeima were segmented, and the areas of cytoplasmic and lysosomal mtKeima were determined.The mitophagy index was calculated as the ratio of the total area of lysosomal mitochondria to the total area of cytoplasmic mitochondria per well.
Transmission electron microscopy.The right myocardium or H9C2 cardiomyocytes were xed in 2.5% glutaraldehyde for 2 h and immersed in 1% osmic acid for 2 h at 4°C.The xed samples were then washed in PBS, dehydrated in a graded series of ethanol.Subsequently, the samples were embedded in Epon 812 (SPI Supplies, West Chester, PA, USA) and placed in a model for polymerization.After the semi thin section was used for positioning, the ultrathin section was made and collected for microstructure analysis, followed by counterstaining with 3% uranyl acetate and 2.7% lead citrate.Next, we observed the sections using a HT7800 TEM (HITACHI, Tokyo, Japan) operating at 100 kV.LC-MS metabolomics analysis.We weighed 60 mg of sample and added 20 µL of 2-chloro-lphenylalanine (0.3 mg / mL, dissolved in methanol) and 0.6 mL of mixed solution (methanol/water = 7/3 (v: v)) into the 1.5 mL EP tube.The samples were homogenized for 2 min and then extracted 30 min by sonication.They were then placed at -20°C for 20 min and centrifuged at 13000 g for 15 min (4℃).LC-HRMS was performed on a Waters UPLC I-class system equipped with a binary solvent delivery manager and a sample manager, coupled with a Waters VION IMS Q-TOF Mass Spectrometer equipped with an electrospray interface (Waters Corporation, Milford, USA).The injection volume was 3.00 µL, and the column temperature was set at 45℃.The mass spectrometric data was collected using a Waters VION IMS Q-TOF Mass Spectrometer equipped with an electrospray ionization source operating in either positive or negative ion mode.The source and desolvation temperatures were set at 120℃ and 500℃, respectively, with a desolvation gas ow of 900 L/h.Centroid data was collected from 50 to 1000 m/z with a scan time of 0.1 s and an interscan delay of 0.02 s over a 13-min analysis time.
iTRAQ proteomics analysis/nanoUHPLC-MS/MS analysis.Lysis buffer (1% SDS, 8 M urea, 1x Protease Inhibitor Cocktail [Roche Ltd.Basel, Switzerland]) was added into the samples and vibrated and milled for 400 s thrice.The samples were then lysed on ice for 30 min and centrifuged at 15000 rpm for 15 min at 4℃.The protein concentration of the supernatant was determined using the BCA protein assay; we then transferred 100 µg of protein/condition into a new Eppendorf tube and adjusted the nal volume to 100 µL with 8 M urea.We added 2 µL of 0.5 M TCEP and incubated the sample at 37℃ for 1 h; next, 4 µL of 1 M iodoacetamide was added to the sample, and the incubation lasted 40 min protected from light at room temperature.Five volumes of -20℃ pre-chilled acetone were then added to precipitate the proteins overnight at -20℃.The precipitates were washed by 1-mL pre-chilled 90% acetone aqueous solution twice and then re-dissolved in 100 µL 100 mM TEAB.Sequence grade modi ed trypsin (Promega, Madison, WI) was added at the ratio of 1:50 (enzyme: protein, weight: weight) to digest the proteins at 37℃ overnight.The peptide mixture was desalted by C18 ZipTip, quanti ed by Pierce™ Quantitative Colorimetric Peptide Assay (23275), and lyophilized by SpeedVac.
The resultant peptide mixture was labeled with iTRAQ 8Plex labeling kit (Sciex) following the manufacturer's instructions.The labeled peptide samples were then pooled and lyophilized in a vacuum concentrator.The peptide mixture was re-dissolved in the buffer A (20 mM ammonium formate in water, pH 10.0, adjusted with ammonium hydroxide), and then fractionated by high pH separation using Ultimate 3000 system (ThermoFisher Scienti c, MA, USA) connected to a reverse-phase column (XBridge C18 column, Waters Corporation, MA, USA).High pH separation was performed using a linear gradient, starting from 5% B to 45% B in 40 min (B: 20 mM ammonium formate in 80% ACN, pH 10.0, adjusted with ammonium hydroxide).The peptides were re-dissolved in 5% ACN aqueous solution containing 0.5% formic acid and analyzed by on-line nanospray LC-MS/MS on Q Exactive™ HF-X coupled to EASY-nLC 1200 system (Thermo Fisher Scienti c, MA, USA).The column ow rate was maintained at 250 nL/min.The electrospray voltage of 2 kV versus the inlet of the mass spectrometer was used.
Bioinformatics data analysis.The UPLC-Q-TOF/MS raw data were analyzed using progenesis QI (Waters CorporationMilford, USA) software.The parameters used were retention time (RT) range 0.5-14.0min, mass range 50-1000 Da, and mass tolerance 0.01 Da.Isotopic peaks were excluded from the analysis, the noise elimination level was set at 10.00, the minimum intensity was set to 15% of base peak intensity, and RT tolerance was set at 0.01 min.The Excel le was obtained with three-dimension data sets including m/z, peak RT and peak intensities; RT-m/z pairs were used as the identi er for each ion.The

Figures
Figures

Figure 10 Graphic
Figure 10