The Calcineurin-TFEB-p62 Pathway Mediates the Activation of Cardiac Macroautophagy by Proteasomal Malfunction - PubMed (original) (raw)

. 2020 Jul 31;127(4):502-518.

doi: 10.1161/CIRCRESAHA.119.316007. Epub 2020 May 5.

Jie Li # 1 2, Nirmal Parajuli # 1, Zongwen Tian # 1 3, Penglong Wu # 1 4, Megan T Lewno 1, Jianqiu Zou 2, Wenjuan Wang 2 4, Lynn Bedford 5, R John Mayer 6, Jing Fang 7, Jinbao Liu 4, Taixing Cui 8, Huabo Su 1 2, Xuejun Wang 1

Affiliations

The Calcineurin-TFEB-p62 Pathway Mediates the Activation of Cardiac Macroautophagy by Proteasomal Malfunction

Bo Pan et al. Circ Res. 2020.

Abstract

Rationale: The ubiquitin-proteasome system (UPS) and the autophagic-lysosomal pathway are pivotal to proteostasis. Targeting these pathways is emerging as an attractive strategy for treating cancer. However, a significant proportion of patients who receive a proteasome inhibitor-containing regime show cardiotoxicity. Moreover, UPS and autophagic-lysosomal pathway defects are implicated in cardiac pathogenesis. Hence, a better understanding of the cross-talk between the 2 catabolic pathways will help advance cardiac pathophysiology and medicine.

Objective: Systemic proteasome inhibition (PSMI) was shown to increase p62/SQSTM1 expression and induce myocardial macroautophagy. Here we investigate how proteasome malfunction activates cardiac autophagic-lysosomal pathway.

Methods and results: Myocardial macroautophagy, TFEB (transcription factor EB) expression and activity, and p62 expression were markedly increased in mice with either cardiomyocyte-restricted ablation of Psmc1 (an essential proteasome subunit gene) or pharmacological PSMI. In cultured cardiomyocytes, PSMI-induced increases in TFEB activation and p62 expression were blunted by pharmacological and genetic calcineurin inhibition and by siRNA-mediated Molcn1 silencing. PSMI induced remarkable increases in myocardial autophagic flux in wild type mice but not p62 null (p62-KO) mice. Bortezomib-induced left ventricular wall thickening and diastolic malfunction was exacerbated by p62 deficiency. In cultured cardiomyocytes from wild type mice but not p62-KO mice, PSMI induced increases in LC3-II flux and the lysosomal removal of ubiquitinated proteins. Myocardial TFEB activation by PSMI as reflected by TFEB nuclear localization and target gene expression was strikingly less in p62-KO mice compared with wild type mice.

Conclusions: (1) The activation of cardiac macroautophagy by proteasomal malfunction is mediated by the Mocln1-calcineurin-TFEB-p62 pathway; (2) p62 unexpectedly exerts a feed-forward effect on TFEB activation by proteasome malfunction; and (3) targeting the Mcoln1 (mucolipin1)-calcineurin-TFEB-p62 pathway may provide new means to intervene cardiac autophagic-lysosomal pathway activation during proteasome malfunction.

Keywords: TFEB protein, rat; autophagy; calcineurin; cardiotoxicity; proteasome inhibitor; sequestosome-1; ubiquitin.

PubMed Disclaimer

Conflict of interest statement

DISCLOSURES

The authors declared there was no conflict of interest.

Figures

Figure 1.

Figure 1.. Psmc1CKO increased myocardial p62, LC3-II, and autophagosomes in mice.

A, Western blot analyses for the indicated proteins in the hearts of CTL (yellow circle; n=5) and Psmc1CKO (red circle; n=6) mice at postnatal day 2 (P2). Ub, ubiquitin conjugates. B and C, GFPdgn transgene was introduced into CTL (n=3) and Psmc1CKO (n=5) mice. Myocardial GFPdgn protein levels in indicated P2 mice were assessed by western blot (B) and quantified (C). D, Immunostaining of ubiquitin (green) and p62 (red) in myocardium sections from P2 mice. Arrowheads mark cardiomyocytes with increased ubiquitinated proteins and p62. Nuclei were counter-stained with DAPI (blue). Bar, 10 μm. E and F, Western blots (E) of myocardial LC3 (n=4 mice for CTL and 3 for Psmc1CKO) and p62 (n=4 mice for CTL and 4 for Psmc1CKO) at P2 and the quantification (F). G and H, GFP-LC3 transgene was crossed into CTL and Psmc1CKO (KO) mice to label autophagosome (green puncta). Shown are confocal fluorescent images (G) of myocardium sections from P2 mice and the quantification of GFP-positive puncta (H) where 4 CTL and 3 KO mice were used and 2 technical repeats per mouse were included. Bar, 10 μm. The precise p values are shown immediately above the pairwise comparison bracket (the same for all figures) and they are from two-sided and unpaired t test (C, F), or nested _t_-tests (H).

Figure 2.

Figure 2.. Genetic inhibition of the proteasome increases p62 and autophagic flux in cardiomyocytes.

A - D, Neonatal mouse ventricular cardiomyocytes (NMVMs) were isolated from neonatal PSMC1f/f (A) or WT (B) mice and infected with Ad-Gal (circle) or Ad-Cre (triangle) for 72 hours. The cells were treated with vehicle (Veh) or Bafilomycin A1 (BFA, 100 nM) for 4 hours before harvest. Shown are representative western blots of the indicated proteins (A, B) and the pooled densitometry data of LC3-II (C) and derived LC3-II flux (D). E - J, Neonatal rat ventricular myocytes (NRVMs) were transfected with siRNAs targeting Psmc1 (siPSMC1; red) or luciferase (siLuci, yellow) for 72 hours. Some of the cells were treated with BFA (100 nM, 2 hours) or Veh for autophagic flux assays (G-J). E-I, Western blots (E, G) and the quantification of p62 (H) and the derived p62 flux (I). Each lane and dot represent a biological repeat; n=4 biological repeats for each group in all quantitative panels. The p values shown are either the precise p values derived from two-sided and unpaired t test (D, I) or the adjusted p values from two way ANOVA followed by Tukey’s tests (C, H) or from multiple t-tests corrected for multiple testing with the Holm-Sidak method (F). J, Representative confocal fluorescent images of cells immunostained for ubiquitin (Ub, green) and LC3 (red).

Figure 3.

Figure 3.. Pharmacological PSMI increases myocardial p62 and activates TFEB in mice.

Mixed sex WT mice at 5 weeks of age were treated with bortezomib (BZM; 10 μg/kg, i.p.; open circle) or vehicle control (60% DMSO in saline; solid circle) for 12 hours. Ventricular myocardium was sampled for protein and RNA analyses. A and B, Western blots (A) of indicated proteins and pooled densitometry data of p62 proteins (B). N= 4 mice/group; two-sided and unpaired _t_-test.. C-E, Western blot (C) of TFEB in the cytoplasmic and nuclear fractions of ventricular myocardial proteins and the densitometry data (D, E). Here in-lane loading control used the stain-free total protein imaging technology (Supplementary Figure 4). GAPDH and histone H3 (H3) serve as cytoplasmic and nuclear proteins marks, respectively. N=3 mice/group; two-sided and unpaired _t_-test. F and G, Representative images of RT-PCR (F) and quantitative real-time PCR (qPCR) data (G) of indicated genes. N=3 mice/group; shown are adjusted p value derived from the multiple t-tests corrected with the Holm-Sidak method. Each lane and each dot represent a unique mouse.

Figure 4.

Figure 4.. Genetic PSMI via Psmc1CKO activates myocardial TFEB in mice.

A and B, Immunostaining of TFEB (green) in the myocardium sections of CTL (yellow circle) and Psmc1CKO (KO; red circle) mice at P2. Representative confocal micrographs (A) and the percentage of TFEB-positive nuclei (arrowheads) were quantified (B). The sections were counterstained with Phalloidin (red) and DAPI (blue), respectively. Bar, 20 μm. C and D, Western blots (C) and the quantification (D) of TFEB in CTL and Psmc1CKO hearts at P2. E, qPCR analyses of indicated genes in CTL and Psmc1CKO hearts. Each lane and each dot represent a unique mouse. The p values shown are from two-sided and unpaired t test (B, D), or multiple t-tests corrected for multiple testing with the Holm-Sidak method (E).

Figure 5.

Figure 5.. PSMI by BZM leads to dephosphorylation and nuclear translocation of TFEB in NRVMs.

NRVMs were treated with BZM (25 nM) or DMSO for 12 or 24 hours before analyses. A, Western blot analyses for TFEB (top). The stain-free total proteins (bottom) on the PVDF membrane were used as loading control. B and C, NRVMs were treated with DSMO or BZM for 12 hours (B) and 24 hours (C), respectively, and subjected to cytosolic and nuclear fractionation, followed by western blot analyses. GAPDH and Histone H3 (H3) were probed as a cytoplasmic and nuclear protein marker, respectively. N = 3 biological repeats/group. D, NRVMs were treated with DMSO or BZM (25 nM) for 12 hours and immunostaind for TFEB (green). The nuclei and F-actin were counterstained with DAPI (blue) and Phalloidin (red), respectively. Representative fluorescence confocal micrographs are shown. Scale bar = 40 μm. E, NRVMs were transfected with siLuci (circle) or siRNA against rat TFEB (siTFEB; triangle) for 48 hours, followed by treatment with BZM (25 nM; red) or vehicle (yellow) for additional 24 hours. Shown are mRNA levels of the indicated genes assessed by qPCR from 3 biological repeats of each group. P values are derived from two way ANOVA followed by Tukey’s test and are all statistically significant after correction for 6 testings (6 genes) with the Benjamini Hochberg procedure.

Figure 6.

Figure 6.. TFEB activation by PSMI in NRVMs is calcineurin- and Mcoln1-dependent.

A and B, NRVMs were treated with cyclosporine A (CsA; 1.5 μM; solid circle) or vehicle control (open circle) for 4 hours and subsequently with BZM (25 nM) for additional 12 hours. Western blot analyses for TFEB in the cytosolic and nuclear fractions (A) and the densitometry data (B). GAPDH and histone H3 (H3) were probed as a cytoplasmic and nuclear protein marker, respectively. C, Knockdown of calcineurin (Cn) Aβ with specific siRNA (siCnAβ) significantly attenuated the activation of TFEB target genes by proteasome inhibition with BZM. NRVMs were transfected with siCnAβ (triangle) or siLuci (circle) for 48 hours and then treated with bortezomib (BZM, 25 nM; solid symbols) or vehicle control (open symbol) for additional 24 hours. The transcripts of the indicated genes were assessed with qPCR. D-F, NRVMs were transfected with siRNAs specific for Mcoln1 (si-Mcoln1) or for luciferase (si-Luc) for 72 hours and treated with BZM (25 nM) for additional 12 hours. The cells were subjected to subcellular protein fractionation or for total RNA extraction. D and E, Western blot analyses for TFEB in the cytosolic and nuclear fractions (D) and the densitometry data (E). F, qPCR analyses for the indicated TFEB target genes. The p values shown are from two-sided and unpaired t test (B, E) or two way ANOVA followed by Tukey’s tests (C, F). All the p values smaller 0.05 in C and F remain statistically significant after correction for multiple-testing (6 genes in parallel) using the Benjamini Hochberg procedure.

Figure 7.

Figure 7.. p62 is required for PSMI to increase autophagic flux in mouse hearts and cardiomyocytes.

A-C, Western blots (A) and the quantification of myocardial LC3-II (B) and the derived LC3-II flux (C). WT and p62KO mice at 6-8 weeks of age were first treated with MG262 (5 μmol/kg, i.p.; solid symbol) or vehicle control (DMSO; open symbol) for 11 hours and subsequently with bafilomycin A1 (BFA, 3 μmol/kg, i.p.; triangle) or DMSO (circle) for additional 1 hour. Ventricular myocardium were collected for the analyses. N=3 mice for each group. D-F, LC3-II flux assays for cardiomyocytes isolated from WT and p62KO mice at postnatal day 2. The cells were treated with bortezomib (BZM; 25 nM) for 6 hours, followed by BFA (25 nM) treatment for additional 6 hours. Shown are western blots (D) of indicated proteins and the quantification of LC3-II (E) and derived LC3-II flux (F) from 3 biological repeats. The stain-free total proteins serve as loading control (L.C.) and a segment of the image is shown. Data were analyzed by two-way (C, F) or three-way (B, E) ANOVA, which all show statistical significance in both treatment effects and interaction among factors; as such they were followed by Tukey’s post hoc multiple comparisons tests to give rise the p values shown in the figure. N = 3 mice for each group.

Figure 8.

Figure 8.. Ablation of p62 (p62KO) attenuates myocardial TFEB activation by PSMI in mice.

WT and p62KO mice were treated with MG262 or vehicle control as described in Figure 7A. Ventricular myocardium was collected for the analyses. A, Confocal micrographs of immunofluorescence staining for TFEB (green) and counter-staining with DAPI (blue) for nuclei and with Phalloidin (red) for F-actin. Bar = 40 μm. B, qPCR analyses for the indicated target genes of TFEB. Shown p values are derived from two way ANOVA followed by Tukey’s test and are all statistically significant after correction for multiple-testing (5 genes) with the Holm-Sidak method. C, An overall model for proteasome malfunction to activate autophagy through the Mcoln1-calcineurin-TFEB-p62 pathway. Proteasome malfunction accumulates and activates calcineurin, which in turn dephosphorylates and activates TFEB; the activation of TFEB increases the expression of Mcoln1, p62 and other genes of the CLEAR network and thereby increases autophagy. Mcoln1 and p62 in turn exert a feed-forward effect on TFEB activation; p62 does so potentially through sequestration of mTORC1 into the aggregates of ubiquitinated proteins and thereby preventing mTORC1 from phosphorylating TFEB. Meanwhile, p62 recruits and condenses ubiquitinated proteins for autophagic degradation. PSMI, proteasome inhibition; CnA, calcineurin; Psmc1KO, cardiomyocyte-restricted knockout of Psmc1; Ub. pro., ubiquitinated proteins.

Comment in

References

    1. Wang X, Pattison JS and Su H. Posttranslational modification and quality control. Circ Res. 2013;112:367–81. - PMC - PubMed
    1. Wang X and Robbins J. Proteasomal and lysosomal protein degradation and heart disease. J Mol Cell Cardiol. 2014;71:16–24. - PMC - PubMed
    1. Wang C and Wang X. The interplay between autophagy and the ubiquitin-proteasome system in cardiac proteotoxicity. Biochim Biophys Acta. 2015;1852:188–94. - PMC - PubMed
    1. Collins GA and Goldberg AL. The Logic of the 26S Proteasome. Cell. 2017;169:792–806. - PMC - PubMed
    1. Day SM. The ubiquitin proteasome system in human cardiomyopathies and heart failure. Am J Physiol Heart Circ Physiol. 2013;304:H1283–93. - PMC - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources