Acute endoplasmic reticulum stress-induced mitochondria respiratory chain damage: The role of activated calpains

Abstract

The induction of acute endoplasmic reticulum (ER) stress damages the electron transport chain (ETC) in cardiac mitochondria. Activation of mitochondria-localized calpain 1 (CPN1) and calpain 2 (CPN2) impairs the ETC in pathological conditions, including aging and ischemia–reperfusion in settings where ER stress is increased. We asked if the activation of calpains causes the damage to the ETC during ER stress. Control littermate and CPNS1 (calpain small regulatory subunit 1) deletion mice were used in the current study. CPNS1 is an essential subunit required to maintain CPN1 and CPN2 activities, and deletion of CPNS1 prevents their activation. Tunicamycin (TUNI, 0.4 mg/kg) was used to induce ER stress in C57BL/6 mice. Cardiac mitochondria were isolated after 72 h of TUNI treatment. ER stress was increased in both control littermate and CPNS1 deletion mice with TUNI treatment. The TUNI treatment activated both cytosolic and mitochondrial CPN1 and 2 (CPN1/2) in control but not in CPNS1 deletion mice. TUNI treatment led to decreased oxidative phosphorylation and complex I activity in control but not in CPNS1 deletion mice compared to vehicle. The contents of complex I subunits, including NDUFV2 and ND5, were decreased in control but not in CPNS1 deletion mice. TUNI treatment also led to decreased oxidation through cytochrome oxidase (COX) only in control mice. Proteomic study showed that subunit 2 of COX was decreased in control but not in CPNS1 deletion mice. Our results provide a direct link between activation of CPN1/2 and complex I and COX damage during acute ER stress.

1 |. INTRODUCTION

The endoplasmic reticulum (ER) plays a crucial role in protein folding, lipid synthesis, and calcium homeostasis.¹ An increase in protein misfolding or disruption of calcium homeostasis inside the ER triggers ER stress.² Although an initial response of ER stress is to restore ER function by decreasing protein synthesis and improving protein folding, prolonged and severe ER stress activates response programs that cause cell injury and death. ER stress is involved in cell injury during many pathological conditions, including aging,^3–5 diabetic disease,⁶ doxorubicin cardiotoxicity,⁷ and heart failure.⁸ The ER and mitochondria are closely connected and communicate through mitochondria associated membranes.^9,10 Thus, ER stress has a direct impact on mitochondrial function.⁹ The induction of acute ER stress impairs the function of the mitochondrial electron transport chain (ETC) in multiple animal models.^2,11,12 The impairment of the ETC not only decreases ATP production but also promotes tissue injury via the production and release of toxic substances, including reactive oxygen species (ROS) and mitochondrial peptides.^13,14 Thus, it is critical to understand the mechanisms whereby ER stress leads to ETC damage to provide the basis for developing mechanistic approaches to protect mitochondrial function in diseases that increase ER stress in the heart.

Calpain 1 (CPN1) and calpain 2 (CPN2) are calcium-dependent cysteine proteases found in both the cytosol and mitochondria.^15–18 The activation of CPN1 and CPN2 (CPN1/2) increases cell injury during cardiac ischemia–reperfusion.^16,19,20 Mitochondrial-localized CPN1/2 (mCPN1/2) activation contributes to mitochondrial damage in diverse pathologies, including aging, ischemia–reperfusion injury, diabetic cardiomyopathy, doxorubicin-induced cardiotoxicity, and heart failure.^17,21–28 Our previous research revealed that the induction of acute ER stress activates CPN1.^29–31 In addition, the induction of ER stress can increase apoptosis by activating calpain 1 (CPN1). Activation of mitochondrial CPN2 (mCPN2) also leads to mitochondrial damage and mitochondrial permeability transition pore (MPTP) opening in isolated rat hearts following ischemia–reperfusion.¹⁷ Activation of CPN2 contributes to cell injury during ischemia–reperfusion and ER stress.^18,32 Overexpression of CPN2 leads to age-dependent dilated cardiomyopathy.³³ Taken together, activation of CPN1/2 contributes to cardiac injury in pathological conditions. We propose that the induction of acute ER stress impairs the ETC by activating mCPN1/2.

Disruption of protein folding increases ER stress.² Tunicamycin (TUNI) is often used to induce ER stress by increasing protein misfolding through direct interference with protein glycosylation.^34,35 Our previous study showed that TUNI treatment led to damage to complex I in cardiac mitochondria isolated from C57BL/6 mice.²⁹ The TUNI treatment also activates both cCPN1/2 and mCPN1/2,²⁹ indicating that activation of CPN1/2 may be involved in mitochondrial injury during acute ER stress. To investigate the causative link between CPN1/2 activation and mitochondrial damage during ER stress, we studied the effects of TUNI-induced ER stress on mitochondrial function in littermate control and CPNS1 (small regulatory subunit one) deletion mice. CPN1 includes one large subunit (78 kD) and one small regulatory subunit (CPNS1).^36,37 Both subunits are required to maintain normal CPN1 activity. Similarly, CPN2 includes a large subunit isoform (80 kD) and CPNS1. Thus, cardiomyocyte-specific inducible CPNS1 deletion mice were utilized to genetically eliminate the activities of both CPN1 and CPN2.³⁶

We hypothesize that genetic deletion of CPNS1 that eliminates the activities of calpain 1 and calpain 2 will protect mitochondria during TUNI-induced acute ER stress. Several complex I subunits are substrates of mitochondrial CPN1 and CPN2.^1,2 Thus, we propose that the elimination of the CPN1/2 activity will preserve complex I activity and composition during ER stress by protecting subunits of complex I. Results from this study will expand our previous findings²⁹ and provide direct evidence that activation of CPN1/2 leads to mitochondrial dysfunction during acute ER stress. In addition, our results can also provide therapeutic guidance to decrease cardiac injury during ER stress through pharmacologic inhibition of CPN1/2.

2 |. MATERIALS AND METHODS

The animal experiments were conducted in accordance with the ethical guidelines outlined in the Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee of our institutions (Committees of Virginia Commonwealth University (VCU) and the Richmond Department of Veterans Affairs Medical Center).

2.1 |. Induction of ER stress in control and CPNS1 deletion mice using tunicamycin (TUNI)

Adult male mice, both control and CPNS1 deletion strains, were employed in this study. CPN1 and CPN2 consist of large subunits and a small regulatory subunit CPNS1. The activity of CPN1/2 is dependent on the presence of both subunits, and the removal of CPNS1 results in the complete elimination of CPN1/2 activity. The floxed CPNS1^PZ/PZ mice were generated in a C57BL/6 background by Dr. Peter Greer at Queen’s University Cancer Research Institute in Kingston, Ontario, Canada.³⁶ Cardiac-specific CPNS1 deletion mice (CPN4^PZ/PZ) were created by crossing the floxed CPNS1^PZ/PZ mice with MHC-MerCreMer mice [B6.FVB(129)-A1cfTg(Myh6-cre/Esr1*)1Jmk/J], which also have a C57BL/6 background and harbor the cardiac-specific, tamoxifen-inducible cre recombinase. The first generation of mice produced CPNS1^PZ/W and CPNS1^PZ/W. cre mice. The CPNS1^PZ/W.cre mice were subsequently crossed with CAPNS1^PZ/PZ mice to generate CPNS1^PZ/PZ. cre mice. Genotyping of CPNS1^PZ/PZ.cre mice was performed using the following PCR primers: P1 (GTC AGG CTA GAT GCC ATG TTC C), P2 (CGA CTA TCC GAG CGC TGC C), and P3 (GTT CAC TTG GAT CTG TCC GGT GCC). The primers used for cre detection were: P4 (ATA TCT CAC GTA CTG ACG GTG GG) and P5 (CTG TTT CAC TAT CCA GGT TAG GG). Tamoxifen treatment [IP injection (1 mg/day), daily for 4 days]³⁸ was administered to CPNS1^PZ/PZ.cre mice to induce cardiac-specific CPNS1 deletion mice. Tamoxifen-treated CPNS1^PZ/PZ mice lacking cre served as controls.

Tunicamycin (TUNI) at a dose of 0.4 mg/kg was administered to C57BL/6 mice in vivo through a single intraperitoneal injection.²⁹ TUNI was initially dissolved in DMSO and then diluted with saline for injection.² DMSO and saline solutions were used for vehicle treatments. After a 72-h period following TUNI or vehicle treatment, mice were anesthetized with pentobarbital sodium (100 mg/kg, i.p.), and their hearts were harvested for mitochondrial isolation.¹⁵ Previous research has demonstrated that this TUNI treatment regimen did not induce cardiac contractile dysfunction, as evidenced by the preserved left ventricular ejection fraction and fractional shortening in control mice.²⁹

It is worth noting that tamoxifen has been shown to inhibit mitochondrial function in a dose-dependent manner,³⁹ and mitochondrial function was also assessed in control mice, both with and without tamoxifen treatment alone.

2.2 |. Isolation of cytosol and mitochondria from a single mouse heart

Heart mitochondria were isolated following a previously established protocol.³ First, the harvested mouse heart was quickly blotted dry, weighed, and minced in cold buffer A. This buffer contained the following composition in mM: 100 KCl, 50 MOPS (3 (N-morpholino) propanesulfonic acid), 1 EGTA, 5 MgSO₄, and 1 ATP. The minced heart tissue was then homogenized using a polytron tissue homogenizer at 10 000 rpm for 2.5 s. The polytron homogenate was centrifuged at 500× g for 10 min to separate the supernatant and pellets. The supernatant was further centrifuged at 100 000× g for 30 min to generate particle-free cytosol. The pellet was resuspended in 3 mL buffer A and incubated with trypsin (5 mg/g tissue) for 15 min at 4°C. The addition of trypsin was to increase mitochondrial protein yield and remove potential cytosolic contamination. An equal volume of buffer B (buffer A + 0.2% bovine serum albumin (BSA)) was added to homogenate at the end of incubation. The resulting mixture was centrifuged at 500× g for 10 min. The supernatant was then centrifuged at 3000× g to pellet mitochondria, which were washed with KME buffer (100 mM KCl, 50 mM MOPS, 0.5 mM EGTA) and centrifuged again at 3000× g to yield the final mitochondrial pellet. Mitochondria were then resuspended in KME buffer for functional study.⁴

2.3 |. Mitochondrial oxidative phosphorylation and enzyme activity

To measure the rate of oxygen consumption in mitochondria, a Clark-type oxygen electrode was used at 30°C following the previously described protocol.⁵ Mitochondria were incubated in oxidative phosphorylation buffer with a composition of 80 mM KCl, 50 mM MOPS, 1 mM EGTA, 5 mM KH₂PO₄, and 1 mg/mL of defatted, dialyzed bovine serum albumin at pH 7.4. The complex I substrates used were glutamate (20 mM) and malate (10 mM) or pyruvate (20 mM) and malate (10 mM). The complex II substrate used was succinate (20 mM) with rotenone (7.5 μM), while TMPD (1 mM)-ascorbate (10 mM) with rotenone (7.5 μM) was used as the complex IV substrate. The enzyme activities of the ETC were determined using previously published methods in detergent-solubilized frozen–thawed mitochondria.⁴

2.4 |. Proteomic study

The trypsin-purified mitochondria were used for proteomic study in that potential cytosolic contamination was removed during trypsin treatment.^4,6 Each mitochondrial sample was fractionated and separated on an SDS-PAGE gel, and the identified band was used for in-gel digestion.⁷ The proteins in the gel band were reduced with dithiothreitol and alkylated with iodoacetamide before digestion with sequencing-grade porcine trypsin (Promega, Madison, WI) in room temperature overnight. The digested peptide samples were analyzed on a Bruker TimsTof Pro2 Q-Tof mass spectrometry system operating in the positive ion mode, coupled with a CaptiveSpray ion source (both from Bruker Daltonik GmbH, Bremen). The HPLC column was a Bruker 15 cm × 75 μm id C18 ReproSil AQ, 1.9 μm, 120 Å reversed-phase capillary chromatography column. One microliter volumes of the extract were injected and the peptides eluted from the column by an acetonitrile/0.1% formic acid gradient at a flow rate of 0.3 μL/min were introduced into the source of the mass spectrometer on-line. The digests were analyzed using a Parallel Accumulation–Serial Fragmentation DDA method to select precursor ions for fragmentation with a TIMS-M S scan followed by 10 PASEF MS/MS scans. The TIMS-MS survey scan was acquired between 0.60 and 1.6 Vs/cm² and 100–1700 m/z with a ramp time of 166 ms. The total cycle time for the PASEF scans was 1.2 s and the MS/MS experiments were performed with a collision energy between 20 eV (0.6 Vs/cm²) and 59 eV (1.6 Vs/cm²). Precursors with 2–5 charges were selected with the target value set to 20 000 a.u and intensity threshold to 2500 a.u. Precursors were dynamically excluded for 0.4 min.

The data were analyzed by using all CID spectra collected in the experiment to search the mouse SwissProt database (17 552 entries, downloaded on 3–23-2022) using the program MSFragger V3.4. The protease was set to full enzymatic activity trypsin with two maximum missed cleavages. Mass tolerances were set at 50 ppm MS1and MS2. Oxidation of methionine and protein N-terminal acetylation were set as variable modifications. Carbamidomethylation of Cis was set as a static modification. Percolator was used for protein/peptide validation with q ≤ .01 as the cutoff threshold. A minimum of two peptides is required for protein identification. FDR rates for peptide and protein were both set at 1%. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE⁸ partner repository with the dataset identifier PXD045666.

2.5 |. Immunoblotting

To analyze the cytosol or mitochondrial samples, samples were solubilized in buffer and denatured at 95°C for 5 min. Next, samples were separated using either 12% or 4%–15% Tris-glycine gels from Bio-Rad, Hercules, CA, and transferred to PVDF membranes using semi-dry transfer. Afterward, the membranes were incubated for 1 h. at room temperature in 5% non-fat dry milk from Bio-Rad mixed in TBST buffer. The membranes were then washed with TBST buffer for 5 min at room temperature. For primary antibodies (Table 1), the membrane was incubated at 4°C overnight, followed by washing with TBST buffer before adding a secondary antibody (HRP-conjugated anti-mouse or anti-rabbit IgG F(ab)2 at 1:10 000 dilution, GE Healthcare Life Sciences, Piscataway, NJ), which was then incubated for 1 h. at room temperature. Second antibody controls are shown in Figure S4. Finally, the blots were developed using ECL Plus Western Blotting Detection Reagents from GE Healthcare Life Sciences, Piscataway, NJ, and the membranes analyzed digitally using Image Lab 6.0 software from Bio-Rad, Hercules, CA.⁹

TABLE 1.

Antibodies used in this study.

Antibody name	Company	Catalog number	Concentration
AIF	Cell signaling	4844	1:1000
BIP	Cell signaling	3177	1:1000
Eif2α	Cell signaling	0968	1:1000
p-eif2α	Cell signaling	9721	1:500
GAPDH	Cell signaling	5174	1:1000
ND5	ThermoFisher	PA5-36600	1:1000
NDFV2	Proteintech	15 301-1-AP	1:1000
Spectrin	Santa Cruz	csc-46 696	1:100
VDAC1	Abcam	ab14715	1:2500

2.6 |. Statistical analyses

Data were expressed as mean ± standard deviation (SD). For comparisons between three or more groups, one-way analysis of variance (ANOVA) was performed following a check for normality and equal variance. When a significant F value was obtained, the Student–Newman–Keuls test for multiple comparisons of means was used. Data that failed the test of normality were compared using a non-parametric Kruskal–Wallis ANOVA on ranks followed by Dunn’s analysis for multiple groups. For comparisons between two groups, an unpaired Student’s t-test was performed. SigmaStat 3.5 (Systat, Richmond, CA) was utilized for the statistical analysis, and a p-value of <.05 was considered statistically significant.

3 |. RESULTS

3.1 |. TUNI treatment increased ER stress in both control and CPNS1 deletion mice

Our previous study showed that 72 h following TUNI treatment, the ER stress was increased in C57BL/6 mice.⁹ Thus, ER stress was assessed in control and CPNS1 deletion mice at this time point. An increase in the contents of Bip¹⁰ and eif-2α¹¹ were used as indicators of ER stress. Compared to vehicle, TUNI treatment increased the content of Bip in both control and deletion mice (Figure 1A,B). In addition, phosphorylated eif-2α in both tamoxifen-treated CPNS1^PZ/PZ control mice and deletion (CPNS1^PZ/PZ cre) mice was also increased in TUNI-treated mice compared to vehicle (Figure 1A,B). TUNI treatment increases ER stress and PERK content in cultured HL-1 cells.¹² Compared to vehicle, TUNI treatment also increased the content of PERK in both control and CPNS1 deletion mice (Figure 1C,D). These results support that TUNI treatment increases the ER stress in both control and deletion mouse hearts.

FIGURE 1.

Tunicamycin (TUNI) treatment increased ER stress in both tamoxifen treated CAPNS1^PZ/PZ control mice and CPNS1 deletion mice. TUNI treatment increased the content of Bip and the phosphorylation of eif2α compared to vehicle in control mice (Panel A and B). In a similar manner, Panel A and B showed that TUNI treatment also increased the contents of Bip and phosphorylated eif2α in deletion mice. Panel C and D showed that TUNI treatment increased the content of PERK in both control and deletion mice (mean ± SD, *p < .05 vs. vehicle). The unpaired Student’s t-test was used for the statistical analysis.

3.2 |. TUNI treatment decreased oxidative phosphorylation only in control mice

The rate of oxidative phosphorylation was measured in isolated mitochondria from control and deletion mouse hearts. Compared to vehicle, TUNI treatment led to a decreased rate of state 3 (ADP-stimulated) respiration in mitochondria isolated from control mice when glutamate + malate was used as complex I substrate (Table 2). TUNI also decreased the rate of state 4 (ADP-limited) respiration in control mice with complex I substrate (Table 2). There were no differences in RCR (respiratory control ratio) between mitochondria from vehicle and TUNI-treated hearts using complex I substrate (Table 2). TUNI treatment also decreased the maximal rate of state 3 respiration (2 mM ADP-stimulated respiration) in control mice using complex I substrate. Uncoupled respiration (0.3 mM DNP-stimulated respiration) was also decreased in mitochondria from TUNI-treated hearts compared to vehicle (Table 2). In contrast, TUNI treatment did not alter respiration in CPNS1 deletion mice, including state 3, maximal state 3 respiration, and uncoupled respiration compared to vehicle in the presence of complex I substrate (Table 2). The rate of state 3, maximal respiration, and uncoupled respiration were higher in TUNI-treated deletion mice compared to TUNI-treated control mice (Table 2). These results indicate that deletion of CPNS1 protects mitochondrial oxidative function during ER stress.

TABLE 2.

Tunicamycin (TUNI) treatment led to decreased oxidative phosphorylation in control but not in CPNS1 deletion mice.

	Control		Deletion
	Vehicle (n = 8)	TUNI (n = 8)	Vehicle (n = 9)	TUNI (n = 9)
Complex I substrates: Glutamate + malate
State 3	237 ± 61	154 ± 42*	263 ± 70	234 ± 61†
State 4	41 ± 4	34 ± 6*	39 ± 8	33 ± 5
RCR	6.0 ± 2.2	4.7 ± 1.4	7.3 ± 2.9	7.1 ± 1.4
ADP (2 mM)	244 ± 49	170 ± 48*	275 ± 88	248 ± 69†
DNP (0.3 mM)	257 ± 55	155 ± 39*	270 ± 63	225 ± 57†
Complex II substrates: Succinate
State 3	700 ± 52	468 ± 149*	719 ± 54	591 ± 149‡
State 4	214 ± 19	143 ± 39*	218 ± 17	204 ± 93
RCR	3.1 ± 0.4	3.2 ± 0.5	3.3 ± 0.2	3.1 ± 0.7
ADP (2 mM)	684 ± 76	419 ± 133*	705 ± 52	556 ± 150‡
DNP (0.3 mM)	621 ± 62	391 ± 50*	649 ± 52	510 ± 136‡
Complex IV substrate: TMPD + ascorbate
ADP (2 mM)	1751 ± 144	1191 ± 408*	1727 ± 153	1468 ± 360

Note: Data are expressed as mean ± SD. Rotenone was used to prevent potential reverse electron flow when succinate and TMPD-ascorbate were used as substrates.

^*p < .05 vs. vehicle-treated control mice.

^†p < .05 vs. TUNI-treated control mice.

^‡p < .05 vs. vehicle-treated deletion mice.

One-way ANOVA and Student–Newman–Keuls test were used for the statistical analysis in data using complex I and complex IV substrates. A non-parametric Kruskal–Wallis one-way ANOVA on ranks followed by Dunn’s analysis were used for multiple groups using succinate as complex II substrates due to failure of either the test of normality or equal distribution.

Compared to vehicle, TUNI treatment led to a decreased rate of state 3 and state 4 respiration in control mitochondria when succinate was used as complex II substrate (Table 2). TUNI treatment also decreased maximal respiration and uncoupled respiration in control mice using complex II substrate (Table 2). The rate of state 3, maximal respiration, and uncoupled respiration were also decreased in deletion mice with TUNI treatment (Table 2). TUNI treatment decreased the maximal rate of respiration in control mitochondria when TMPD + ascorbate was used as complex IV substrate (Table 2). TUNI treatment did not decrease the rate of TMPD oxidation in deletion mice (Table 2).

3.3 |. TUNI treatment decreased complex I activity in control mice

In control mice, TUNI treatment led to decreased complex I activity compared to vehicle (Figure 2A). The activity of NADH dehydrogenase (NFR) was also decreased in TUNI-treated mice (Figure 2B). In contrast, TUNI treatment did not alter the activities of complex I or NFR in deletion mice (Figure 2A,B). The NFR activity in mitochondria from TUNI-treated deletion mice was higher than that in mitochondria from TUNI-treated control mice (Figure 2B). There was no difference in citrate synthase activity in mitochondria between vehicle and TUNI-treated hearts in both control and deletion mice (Figure 2C). Compared to vehicle, TUNI treatment did not alter complex II activity (mU/mg) in control [mean ± SD, 384 ± 138 (vehicle n = 8) vs. 302 ± 140 (TUNI n = 8) p = NS] or deletion mice [mean ± SD, 429 ± 170 (vehicle n = 9) vs. 390 ± 96 (TUNI n = 9) p = NS, one-way ANOVA test].

FIGURE 2.

TUNI treatment led to decreased complex I activity in tamoxifen-treated CPNS1^PZ/PZ control mice. Panel A showed that TUNI treatment decreased complex I activity only in control mice but not in CPNS1 deletion mice compared to vehicle. TUNI treatment also decreased NFR activity in control mice but not in deletion mice (Panel B). Panel C showed that TUNI treatment did not alter citrate synthase activity either in control or in deletion mice. Mean ± SD, *p < .05 vs. vehicle. ^†p < .05 vs. TUNI-treated control mice. N = 8–9 in each group.) A non-parametric Kruskal–Wallis one-way ANOVA on ranks followed by Dunn’s analysis were used for complex I activity analysis due to failure of the test of normality. One-way ANOVA and Student–Newman–Keuls test were used for the statistical analysis in NFR activity analysis and citrate synthase data analysis.

3.4 |. Tamoxifen treatment alone did not alter mitochondrial function in control mice

Enzyme activities of respiratory chain complexes were measured in isolated mitochondria from control mice with or without tamoxifen treatment. Tamoxifen treatment alone also did not alter the activities of complex I, NFR, II, III, and citrate synthase compared to vehicle (Table 3). These results indicate that the currently used dose of tamoxifen does not markedly affect mitochondrial enzyme activities. Thus, the decreased complex I activity in control was due to TUNI-induced ER stress but not to tamoxifen treatment.

TABLE 3.

Tamoxifen treatment alone did not alter enzyme activities of mitochondria respiratory chain in control mice.

Enzyme activity (mU/mg)	Vehicle (n = 5)	Tamoxifen (n = 4)
Complex I	784 ± 103	887 ± 61
NFR	2543 ± 514	2507 ± 191
Complex II	438 ± 84	293 ± 159
Complex III	12 315 ± 562	11 403 ± 695
Citrate synthase	3324 ± 346	2984 ± 241

Note: Data are expressed as mean ± SD. p = NS. The unpaired Student’s t-test was used for the statistical analysis.

3.5 |. TUNI treatment activated cytosolic and mitochondrial calpain1/2 in control mice

TUNI treatment led to activation of cCPN1 and mCPN1 in C57BL/6 mice.⁹ The in-situ activation of both cCPN1/2 and mCPN1/2 was assessed in control and CPNS1 deletion mice. Spectrin is a substrate of cCPN1/2. A decrease in spectrin content or an increase in cleaved spectrin content was used as a biomarker of cCPN1^1,4,13 and cCPN2 activation.¹⁴ TUNI treatment did not alter the spectrin content (250 KD) in control mice compared to vehicle (Figure 3A). However, the contents of cleaved spectrin (150 KD and 90 KD) were increased in TUNI-treated control mice compared to vehicle (Figure 3A). In contrast, TUNI treatment did not increase the cleavage of spectrin in CPNS1 deletion mice compared to vehicle (Figure 3B). These results indicate that TUNI treatment increased the functional activity of cCPN1/2 only in control mice.

FIGURE 3.

TUNI treatment activated cytosolic CPN1/2 in tamoxifen-treated CPNS1 ^PZ/PZ control mice. TUNI treatment did not alter the content of full length spectrin compared to vehicle in control mice (Panel A). However, TUNI increased the content of cleaved spectrin at 150 KD and 75 KD (Panel A). In contrast, TUNI did not increase the cleavage of spectrin in CPNS1 deletion mice (Panel B). These results show that TUNI treatment activates cytosolic CPN1/2 only in control mice. Data are expressed as mean ± SD; unpaired Student’s t-test was used for the statistical analysis. *p < .05 vs. vehicle. N = 7 in each group.

Apoptosis inducing factor (AIF) is a nuclear-encoded protein that is imported into mitochondria as a 67-KD precursor. Mature AIF (62 KD) is formed following the removal of the mitochondrial leader sequence in the precursor. Thus, a decrease in mature AIF (62 KD) or an increase in cleaved AIF (57 KD) indicates mCPN1 activation.^15–17 TUNI treatment decreased the content of mature AIF in mitochondria isolated from control mice compared to vehicle (Figure 4A). However, TUNI treatment did not alter the mature AIF content in deletion mice compared to vehicle (Figure 4B). Interestingly, activation of mCPN2 also plays a role in the release of cleaved AIF from mitochondria.¹⁸ These results indicate that TUNI treatment led to mCPN1/2 activation only in control mice.

FIGURE 4.

TUNI treatment activated mitochondrial calpain 1/2 in tamoxifen-treated CPNS1 ^PZ/PZ control mice. The TUNI treatment led to decreased AIF content in mitochondria from control mice compared to vehicle (Panel A), supporting that TUNI treatment activated mitochondrial CPN1. In contrast, TUNI treatment did not alter AIF content in CPNS1 deletion mice (Panel B). VDAC1 (unchanged by CPNS1 deletion by proteomic analysis) was used as protein loading control. These results support the fact that removing CPNS1 effectively prevents mitochondrial CPN1/2 activation during ER stress. Data are expressed as mean ± SD; *p < .05 vs. vehicle. N = 5 in each group. The unpaired Student’s t-test was used for the statistical analysis.

3.6 |. TUNI treatment altered the contents of metabolic enzymes in control mice

Proteomic analysis was used to identify the deficient proteins in mitochondria from TUNI-treated tamoxifen-treated CPNS1^PZ/PZ control and CPNS1 deletion mice. Compared to vehicle, TUNI treatment led to decreased contents of ATP synthase subunit a, cytochrome oxidase (COX) subunit 2, and ferrochelatase in control mice (Table 4).

TABLE 4.

ER stress induced by tunicamycin (TUNI) treatment led to decreased mitochondrial protein contents in control but not in deletion mice.

		Gene	Control		Deletion
Protein	Accession	ID	LFQ ratios TUNI/Vehicle	p-Value	TUNI/Vehicle LFQ ratios	p-Value
ATP synthase subunit a	P00848	Mtatp6	0.00	.062	1.47	.677
Cytochrome oxidase subunit 2	P00405	Mtco2	0.58	.018	1.23	.519
Ferrochelatase	P22315	Fech	0.23	.009	0.88	.835
NADH–ubiquinone oxidoreductase chain 5	P03921	Mtnd5	0.62	.043	0.88	.460
NADH dehydrogenase flavoprotein 2	Q9D6J6	Ndufv2	0.74	.002	0.74	.159
Voltage-dependent anion-selective channel protein 2	Q60930	VDAC2	0.59	.033	0.75	.089
Voltage-dependent anion-selective channel protein 3	Q60931	VDAC3	0.53	.011	0.86	.440

Note: Data are expressed as LFQ ratio.

The contents of complex I subunits, including NADH–ubiquinone oxidoreductase chain 5 (ND5) and NADH dehydrogenase flavoprotein 2 (NDUFV2), were also decreased in TUNI-treated control mice ( Table 4). The contents of VDAC2 and VDAC3, but not VDAC1, were decreased in control mice with TUNI treatment (Table 4). TUNI treatment did not alter the content of any of these proteins in deletion mice compared to vehicle (Table 4).

Interestingly, TUNI treatment also led to an increased content of a few proteins, including ATPase inhibitor and ATP synthase subunit epsilon compared to vehicle (Table 5). Again, these protein contents were not changed in TUNI-treated deletion mice (Table 5).

TABLE 5.

ER stress induced by tunicamycin (TUNI) treatment led to an increase in select mitochondrial protein contents in control but not in deletion mice.

		Gene	Control		Deletion
Protein	Accession	ID	LFQ ratios TUNI/Vehicle	p-Value	TUNI/Vehicle LFQ ratios	p-Value
ATPase inhibitor	O35143	Atp5if1	5.87	.055	1.22	.827
ATP synthase subunit epsilon	P56382	Atp5f1e	4.13	.072	1.12	.875

Note: Data are expressed as LFQ ratio.

3.7 |. TUNI treatment led to decreased complex I subunits in control mice

Immunoblotting was used to confirm the changes in the content of complex I subunits shown in the proteomic study. TUNI treatment led to a decreased content of NDUFV2 (Figure 5A) in mitochondria isolated from control mice compared to vehicle treatment. TUNI treatment did not alter the content of NDUFV2 (Figure 5B) in deletion mice. Compared to vehicle, TUNI treatment led to decreased ND5 (NADH dehydrogenase 5) content in tamoxifen-treated CPNS1^PZ/PZ control mice (Figure 5C). In contrast, the content of ND5 was not altered in TUNI-treated CPNS1 deletion mice (Figure 5D). These results verified the proteomic results that TUNI treatment led to decreased contents of complex I subunits in control mice.

FIGURE 5.

Acute ER stress induced by tunicamycin (TUNI) treatment led to the degradation of complex I subunits in tamoxifen-treated CPNS1^PZ/PZ control mice. NDUFV2 is a nuclear gene-encoded complex I subunit. Compared to vehicle, TUNI treatment led to a decreased content of NDUFV2 in control mice (Panel A) but not in CPNS1 deletion mice (Panel B). VDAC1 (unchanged by CPNS1 deletion by proteomic analysis) was used as protein loading control. ND5 is a mitochondrial gene-encoded complex I subunit. TUNI treatment decreased the content of ND5 in control mice (Panel C) but not in deletion mice (Panel D). These results indicate that ER stress leads to the degradation of both nuclear and mitochondrial DNA-encoded subunits of complex I. Ferrochelatase (FERR) contributes a key role in incorporating ferrous iron into the heme protein. TUNI treatment decreased the content of FERR in control mice (Panel E) but not in deletion mice (Panel F). Subunit 4 of COX was used as protein loading control (mean ± SD, *p < .05 vs. vehicle. N = 5 in each group). The unpaired Student’s t-test was used for the statistical analysis.

TUNI treatment led to a decreased content of ferrochelatase (Table 4) in control but not in deletion mitochondria compared to vehicle treatment. This result was also confirmed by immunoblotting (Figure 5E,F).

3.8 |. TUNI treatment increased the release of cytochrome c from mitochondria in control mice

Immunoblotting was used to detect cytochrome c in purified cytosol. TUNI treatment led to an increased content of cytochrome c in cytosol in tamoxifen-treated CPNS1 ^PZ/PZ control mice compared to vehicle (Figure 6A). TUNI treatment did not increase the release of cytochrome c into cytosol in deletion mice (Figure 6B). These results indicate that prevention of CPN1/2 activation prevents the loss of cytochrome c from mitochondria during acute ER stress.

FIGURE 6.

Acute ER stress induced by tunicamycin (TUNI) treatment increased the release of cytochrome c from mitochondria in tamoxifen-treated CPNS1 ^PZ/PZ control mice. Compared to vehicle, TUNI treatment increased the content of cytochrome c in cytosol in control mice (Panel A) but not in CPNS1 deletion mice (Panel B). GAPDH was used as protein loading control. In addition, total protein content was also used as protein loading control in both control and deletion mice. These results indicate that ER stress leads to the loss of cytochrome c from mitochondria (mean ± SD, *p < .05 vs. vehicle. N = 7 in each group). The unpaired Student’s t-test was used for the statistical analysis.

4 |. DISCUSSION

The induction of acute ER stress damages the ETC in C57BL/6 mice.^1,9,19 In the present study, we find that TUNI-induced ER stress damages the ETC in tamoxifen-treated CPNS1^PZ/PZ control mice but not in CPNS1 cardiac-specific deletion mice. TUNI treatment activates both cCPN1/2 and mCPN1/2 in control mice. The induction of acute ER stress by TUNI treatment resulted in a decrease in oxidative phosphorylation with complex I substrates (Table 2) and a corresponding decrease in the enzyme activity of complex I (Figure 2). The decrease in complex I enzyme activity was due to a decrease in the content of complex I subunits observed in control but not in deletion mice (Table 4, Figure 5). These results provide solid evidence that acute ER stress damages complex I through degradation of its subunits via the activation of mCPN1/2.

Our previous study showed that 72 h after a one-time TUNI treatment that ER stress was increased accompanied by mitochondrial dysfunction and damage to the ETC in C57BL/6 mice.⁹ Therefore, TUNI was used in the current study to induce ER stress in control and CPNS1 deletion mice to test the hypothesis that CPN1/2 was a critical effector of ER stress-mediated mitochondrial dysfunction. The TUNI treatment increased the contents of BIP and PERK and the phosphorylation of eIF2α in control mice, supporting that TUNI treatment increased the ER stress. TUNI treatment increased the ER stress in a similar manner in CPNS1 deletion mice as shown by the increased contents of BIP and PERK and the phosphorylation of eIF2α. TUNI treatment led to increased cleavage of spectrin in control mice, supporting that ER stress activates cytosolic CPN1/2. In contrast, the cleavage of spectrin was not increased in TUNI-treated CPNS1 deletion mice, supporting that the genetic downregulation of CPNS1 prevents CPN1/2 activation during the ER stress. TUNI treatment led to increased ER stress without activation of CPN1/2 in CPNS1 deletion mice, indicating that TUNI-induced ER stress is not dependent on the CPN1/2 activity.

ER stress contributes to mitochondrial dysfunction in pathological conditions, including aging, heart failure, and chemotherapy.^20,21 The induction of acute ER stress using TUNI damages the ETC, especially in complex I, in C57BL/6 mice. Complex I is the initial respiratory complex in the ETC and contributes a major role in ROS production and the increased susceptibility to MPTP opening.^22,23 In the current study, we found that the induction of ER stress by TUNI treatment led to decreased complex I activity in control mice but not in CPNS1 deletion mice. Tamoxifen treatment alone did not alter complex I activity compared to vehicle, supporting that the decreased complex I activity in control mice was due to TUNI treatment rather than tamoxifen. TUNI treatment also activated mitochondrial CPN1/2 in the control mice. As expected, mitochondrial CPN1/2 was not activated in the CPNS1 deletion mice. These results provide solid evidence that acute ER stress damages complex I by activating mitochondrial CPN1/2, leading to the subsequent cleavage of select subunits.

Complex I includes 45 subunits, and most of the subunits are nuclear gene encoded.²⁴ Complex I contains seven mitochondrial gene-encoded subunits, many of which are embedded in the mitochondrial inner membrane, functioning in proton translocation across the inner membrane or in the catalytic sequence of electron flow through the complex.²⁴ Complex I has a peripheral arm functioning in NADH oxidation and subsequent electron transfer through complex I.²⁴ In the current study, we found that ER stress not only decreased complex I activity but also decreased NFR activity.⁵ The current proteomic study identified that ER stress led to a decrease in the content of NDUFV2, which is one of the subunits belonging to NFR. The decreased NDUFV2 content in control mice identified by proteomic analysis (Table 4) was further verified by immunoblotting (Figure 5). These results strongly support that ER stress leads to decreased NFR activity due to loss of a subunit critical for enzyme activity. ER stress did not alter the activities of complex I or NFR in deletion mice. In addition, NDUFV2 content was not decreased following the induction of ER stress by TUNI treatment in deletion mice. These results indicate that CPN1/2-mediated NDUFV2 cleavage contributes to the decrease in complex I enzyme activity that occurs during acute ER stress.

NADH dehydrogenase subunit 5 (ND5) and subunit 6 (ND6) are mitochondrial gene-encoded subunits located in the membrane arm of the complex I.²⁵ ND5 and ND6 also play a key role in complex I subunit assembly. Mutation of the ND5 gene impairs complex I assembly that contributes to Leigh syndrome.^25,26 Thus, although ND5 is not directly involved in NADH oxidation, ND5 deficiency can also decrease complex I activity by affecting complex I assembly. Our study found that TUNI treatment led to decreased ND5 content in control mice.²⁵ Elimination of CPNS1 in deletion mice protects ND5 content in TUNI-treated mice, indicating that activation of CPN1/2 contributes to ND5 degradation during the ER stress.

The induction of acute ER stress using TUNI leads to decreased complex II activity in SH-SY5Y cells.²⁷ TUNI treatment leads to decreased succinate oxidation in C57BL/6 mice.⁹ The current study also showed that TUNI treatment decreased succinate oxidation in control mice, supporting that acute ER stress results in potential complex II defects. Interestingly, complex II activity was not decreased in TUNI-treated control mice, suggesting that the decreased succinate oxidation is due to potential defects in the distal ETC, including complex III or complex IV.⁵ Interestingly, succinate oxidation was also decreased in CPNS1 deletion mice, indicating that ER stress-induced complex II defects are not mCPN1/2 dependent.

TUNI treatment also led to decreased oxidation through COX (cytochrome oxidase, complex IV) in tamoxifen-treated CPNS1^PZ/PZ control mice. Interestingly, TUNI treatment did not markedly decrease the TMPD oxidation in deletion mice. TMPD is a substrate that donates electrons to COX via cytochrome c.⁵ A decrease in TMPD oxidation can be due to the loss of cytochrome c or a defect in COX. TUNI treatment did increase the loss of cytochrome c from mitochondria in control and deletion mice (Figure 6), suggesting that ER stress contributes to decreased TMPD oxidation through a release of cytochrome c from mitochondria into the cytosol.⁹ The current proteomic study showed that TUNI treatment decreased the contents of ferrochelatase and subunit 2 of COX (COX2) in control mice. Ferrochelatase is a key enzyme to incorporate ferrous iron into the porphyrin ring for heme formation.²⁸ COX includes two hemes (a and aa₃), iron centers bound to subunits.²⁹ A decrease in ferrochelatase activity will affect COX by impairing its subunit I.²⁹ Thus, TUNI treatment may decrease COX activity through degradation of the ferrochelatase or COX2. The genetic deletion of CPNS1 protects the ferrochelatase and subunit 2 of COX and indicates that ferrochelatase and COX2 are substrates of mCPN1/2.

ATP synthase subunit a (ATP6) is a mitochondrial gene-encoded subunit of complex V.³⁰ Mutation in the ATP6 gene leads to mitochondrial dysfunction and dilated cardiomyopathy.³⁰ A decrease in ATP6 content contributes to ATP synthase dysfunction in diabetic hearts.³¹ In the current study, we found that TUNI treatment led to decreased ATP6 content in a CPN1/2-dependent manner. ND5, COX2, and ATP6 are mitochondrial DNA-encoded proteins.³² Voltage-dependent anion channels (VDACs) are key proteins located in the mitochondrial outer membrane.³³ There are three isoforms of VDAC, including VDAC1, VDAC2, and VDAC3.³³ The VDACs play an important role in the transport of adenine nucleotides, NADH, and other metabolites across the outer membrane. VDACs may also contribute to cell injury by interacting with pro-apoptotic bcl-2 family members.³³ The genetic deletion of VDAC1 or VDAC3 leads to mitochondrial dysfunction, and elimination of VDAC2 leads to embryonic lethality in mice.^33,34 Cardiac-specific knockout of VDAC2 results in dilated cardiomyopathy through disruption of calcium homeostasis.³⁵ TUNI treatment led to a decrease in the contents of VDAC2 and VDAC3 in littermate control mice, suggesting that ER stress can impair mitochondrial function through degradation of the VDACs. Equally important, VDAC2 and VDAC3 are identified as likely calpain cleavage targets.

TUNI treatment also led to the increased contents of ATP synthase subunits, including ATP5IF1 (ATPase inhibitory factor 1), ATP5F1ε (ATP synthase epsilon), and ATP5F1δ (ATP synthase delta). Activation of mCPN1/2 contributes to ATP synthase damage.³⁶ ATP5IF1 functions in preventing ATP hydrolysis by binding to the catalytic domain of the ATP synthase.^37,40 ATP5F1ε increases ATP hydrolysis by interacting with F0 subunits, whereas it inhibits ATP hydrolysis by interacting with F1 subunits.^38,39 The ATP synthase delta contributes to the stability of the ATP catalytic sector.⁴¹ An increase in these protein contents during acute ER stress may be an adaptive reaction to preserve energy content to support cardiac function in the presence of calpain-induced ETC defects. The adaptive reaction is decreased in deletion mice due to the protection of mitochondrial function during acute ER stress by the downregulation of mCPN1/2.

There are several limitations in the current study. Our previous study showed that TUNI treatment does not markedly impair cardiac function in C57BL/6 mice.⁹ Thus, cardiac function was not measured in the current study. CHOP is a commonly used indicator of ER stress. However, tamoxifen treatment alone has been shown to increase the CHOP expression.⁴² We found that in the current model tamoxifen treatment alone did not lead to increased contents of CHOP and cleaved ATF6 (Figure S1). Interestingly, TUNI treatment also did not increase the contents of CHOP and cleaved ATF6 in control and deletion mice (Figure S2). Thus, we measured other markers of the ER stress and used Bip and phosphorylated eif2α as indicators of the increased ER stress in TUNI-treated mice. Proteomic study showed that TUNI treatment led to decreased content of subunit 2 of COX in control mice. However, immunoblotting showed that TUNI treatment did not decrease the content of subunit 2 of COX in control mice (Figure S3). This may be due to the quality of the antibody used. Calpain 10 is also a ubiquitous calpain located in the mitochondria.⁴³ Activation of calpain 10 contributes to complex I damage in renal mitochondria.⁴⁴ Calpain 10 is activated during ER stress.⁴⁵ However, calpain 10 is an atypical calpain that does not dimerize with the small regulatory subunit due to lack of EF-hands in domain IV of the large subunit.⁴³ Thus, CPNS1 deletion will not eliminate calpain 10 activity. The activation of calpain 10 is less likely to induce complex I damage in cardiac mitochondria during the acute ER stress of TUNI treatment since the complex I damage was clearly protected by deletion of CPNS1.

In cardiac myocytes, sarcoplasmic reticulum (SR) contributes an important role in regulating intracellular calcium. The SR is also a network of membranes similar to the ER.⁴⁶ The SR surrounds myofilaments and functions in contractile calcium handling by releasing calcium through ryanodine receptor and reuptake through SERCA in cardiac myocytes.⁴⁶ The SR is connected with the peri-nuclear ER in cardiac myocytes, and chaperones for protein synthesis and folding including Grp75 are found in the SR.^46,47 Therefore, protein synthesis may occur in the SR within cardiac myocytes. SR dysfunction contributes to cell injury and the development of heart failure.⁴⁸ Since SR and ER are connected in cardiac myocytes,^46,47 the TUNI treatment may affect both ER and SR. Thus, TUNI treatment may activate the CPN1/2 by impairing the SR function.

5 |. CONCLUSION

The activation of CPN1/2 plays a central role in mitochondrial damage during ER stress. The current mechanistic studies provide substantial insight into the mechanism of ER stress-induced impairment of mitochondrial function in the aging heart^7,16,49 and are likely to be relevant to other cardiac disease states as well, including ischemia–reperfusion¹⁷ and especially heart failure⁵⁰ and doxorubicin cardiotoxicity.⁵¹ Potential strategies to attenuate the ER stress or the direct inhibition of CPN1/2 are promising approaches to protect mitochondria in pathological conditions that lead to increased ER stress, including aging, ischemia–reperfusion, and heart failure.

Abbreviations:

AIF

apoptosis inducing factor

ATP5F1δ

ATP synthase delta

ATP5F1ε

ATP synthase epsilon

ATP5IF1

ATPase inhibitory factor 1

ATP6

ATP synthase subunit a

Bip

78-kDa glucose-regulated protein

BSA

bovine serum albumin

CHOP

C/EBP homologous protein

COX2

subunit 2 of cytochrome oxidase

CPN1

calpain 1

CPN1/2

calpain 1 and calpain 2

CPN2

calpain 2

CPNS1

calpain small regulatory subunit 1

CPNS1PZ/PZ

floxed CPNS1 mice

CPNS1PZ/PZ.cre

CPNS1 deletion mice

DNP

dinitrophenol

ER

endoplasmic reticulum

ETC

electron transport chain

KME

100 mM KCl, 50 mM MOPS, 0.5 mM EGTA

mCPN1

mitochondrial calpain 1

mCPN1/2

mitochondrial-localized CPN1/2

mCPN2

mitochondrial CPN2

MOPS

3 (N-morpholino) propanesulfonic acid

MPTP

mitochondrial permeability transition pore

ND5

NADH dehydrogenase subunit 5

ND6

NADH dehydrogenase subunit 6

NDUFV2

ubiquinone oxidoreductase core subunit V2

NFR

NADH-ferricyanide reductase

RCR

respiratory control ratio

ROS

reactive oxygen species

SERCA

sarcoplasmic/endoplasmic reticulum Ca2+-ATPase

SR

sarcoplasmic reticulum

State 3

ADP-stimulated respiration

State 4

ADP-limited respiration

TMPD

N,N, N′, N′-tetramethylphenylene-1,4-diamine

TUNI

tunicamycin

VDAC1

voltage dependent anion channel 1

VDAC2

voltage dependent anion channel 2

VDAC3

voltage dependent anion channel 3

VDACs

voltage dependent anion channels

DATA AVAILABILITY STATEMENT

Not applicable.