Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 May 25;15(5):363.
doi: 10.1038/s41419-024-06750-2.

Neuronal autosis is Na+/K+-ATPase alpha 3-dependent and involved in hypoxic-ischemic neuronal death

Pauline Depierre #  1 , Vanessa Ginet #  1   2 , Anita C Truttmann  2 , Julien Puyal  3   4
Affiliations

Neuronal autosis is Na+/K+-ATPase alpha 3-dependent and involved in hypoxic-ischemic neuronal death

Pauline Depierre et al. Cell Death Dis. .

Abstract

Macroautophagy (hereafter called autophagy) is an essential physiological process of degradation of organelles and long-lived proteins. The discovery of autosis, a Na+/K+-ATPase (ATP1)-dependent type of autophagic cell death with specific morphological and biochemical features, has strongly contributed to the acceptance of a pro-death role of autophagy. However, the occurrence and relevance of autosis in neurons has never been clearly investigated, whereas we previously provided evidence that autophagy mechanisms could be involved in neuronal death in different in vitro and in vivo rodent models of hypoxia-ischemia (HI) and that morphological features of autosis were observed in dying neurons following rat perinatal cerebral HI. In the present study, we demonstrated that neuronal autosis could occur in primary cortical neurons using two different stimulations enhancing autophagy flux and neuronal death: a neurotoxic concentration of Tat-BECN1 (an autophagy-inducing peptide) and a hypoxic/excitotoxic stimulus (mimicking neuronal death induced by cerebral HI). Both stimulations induce autophagic neuronal death (dependent on canonical autophagic genes and independent on apoptotic, necroptotic or ferroptotic pathways) with all morphological and biochemical (ATP1a-dependent) features of autosis. However, we demonstrated that autosis is not dependent on the ubiquitous subunit ATP1a1 in neurons, as in dividing cell types, but on the neuronal specific ATP1a3 subunit. We also provided evidence that, in different in vitro and in vivo models where autosis is induced, ATP1a3-BECN1 interaction is increased and prevented by cardiac glycosides treatment. Interestingly, an increase in ATP1a3-BECN1 interaction is also detected in dying neurons in the autoptic brains of human newborns with severe hypoxic-ischemic encephalopathy (HIE). Altogether, these results suggest that ATP1a3-BECN1-dependent autosis could play an important role in neuronal death in HI conditions, paving the way for the development of new neuroprotective strategies in hypoxic-ischemic conditions including in severe case of human HIE.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1. High concentrations of Tat-BECN1 induce neuronal death and increase autophagy flux in primary cortical neurons.
A Tat-BECN1 increases the percentage of propidium iodide (PI)-positive nuclei at both 5 and 10 µM after 6 h of treatment compared to non-treated neurons. Tat-sc has a slight toxic effect from 10 µM. Dunn’s multiple comparisons test, compared to non-treated neurons. n ≥ 36 per condition. 4 independent experiments. B The percentage of PI-positive nuclei increases from 3 h in the presence of 5 µM Tat-BECN1 as illustrated in representative images of PI stainings. n ≥ 48 per condition. Tukey’s multiple comparisons test. 4 independent experiments. ***p < 0.001. Scale bar = 100 µm. C Representative confocal microscopy images of primary cortical neurons transfected with the RFP-GFP-LC3 plasmid in Tat-sc and Tat-BECN1 (5 μM) conditions at 1.5 h, 3 h and 6 h. Scale bar = 10 µm. The number of both GFP+ - RFP+ (yellow) and RFP+-GFP− (red) -LC3-positive dots and the total number of LC3-positive dots significantly increase at all time points with Tat-BECN1 treatment. Dunn’s multiple comparisons test, n ≥ 25 per condition; 5 independent experiments. D Representative immunoblots and corresponding quantifications of LC3-II (Unpaired t-test) and SQSTM1/p62 (Mann–Whitney test) after 6 h of Tat-BECN1 (5 μM) treatment (n ≥ 8 per condition, 4 independent experiments) and E after 3 h with or without the lysosomal enzymes inhibitors cocktail E64d/PepstatinA (E64d/PepA) treatment. Both SQSTM1/p62 and LC3-II expression levels are significantly increased by E64d/PepA in basal and Tat-BECN1 conditions. n ≥ 9 per condition. Dunn’s multiple comparisons test. 3 independent experiments. Values are mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001; ns not significant.
Fig. 2
Fig. 2. Tat-BECN1-induced neuronal death in primary cortical neurons is autophagy-dependent and is not involving apoptosis, necroptosis and ferroptosis.
Downregulation of (A) ATG5 (n ≥ 47 per condition, 5 independent experiments), (B) BECN1 (n ≥ 36 per condition, 3 independent experiments) and (C) ATG7 (n ≥ 42 per condition, 4 independent experiments) reduces neuronal death as shown by a decrease in the percentage of propidium iodide (PI) positive nuclei. Dunn’s multiple comparisons test. D The pharmacological pan-caspase inhibitor Q-VD-Oph (25 µM) does not reduce the percentage of PI positive nuclei after 6 h of Tat-BECN1 treatment (n ≥ 30 per condition). E BCL2 overexpression (BCL2 OE) has no effect on Tat-BECN1 increased percentage of PI-positive nuclei after Tat-BECN1 treatment (n ≥ 24 per condition). F Pharmacological inhibition of necroptosis with Necrostatin 1 (Nec-1, 30 µM) does not reduce Tat-BECN1-induced increase percentage of PI-positive nuclei (n ≥ 46 per condition). G Lentiviral-mediated downregulation of RIPK3 (n ≥ 24 per condition) or H of MLKL (n ≥ 36 per condition) has no effect on the percentage of PI-positive nuclei. I Pharmacological inhibition of ferroptosis with Ferrostatin-1 (Fer-1, 10 µM), does not reduce the percentage of PI-positive nuclei (n ≥ 24 per condition; Dunn’s multiple comparisons test, 3 independent experiments). J Lentiviral-mediated overexpression of GPX4 (GPX4 OE) (n ≥ 24 per condition) or K knockdown of ACSL4 (n ≥ 24 per condition) have not effect on PI-positive nuclei number. Dunn’s multiple comparisons test, 3 independent experiments. Values are mean ± SEM. ***p < 0.001; ns: not significant.
Fig. 3
Fig. 3. Tat-BECN1-induced neuronal death has morphological and biochemical features of autosis.
A Representative electron micrographs of primary cortical neurons treated with Tat-scrambled (Tat-sc) or Tat–BECN1 for 3 h. Tat-BECN1 treatment induces typical morphological features of (iii–iv) phase 1a autosis including presence of convoluted nuclei accompanied by endoplasmic reticulum dilatation and fragmentation, and of numerous autophagosomes, autolysosomes and empty vesicles; (v–vi) phase 1b autosis with restricted swollen perinuclear space (PNS) regions (white arrows) containing lumps of cytosol and (vii–x) phase 2 autosis, as shown for different dying neurons, with the presence of a focal ballooning of the (empty) perinuclear space (PNS) associated with a concavity of the nucleus, a swelling of almost all organelles and loss of plasma membrane integrity (black arrows). These morphological features are not observed in Tat-sc treated neurons (i, ii). N: nucleus, ONM: outer nuclear membrane; INM: inner nuclear membrane; PNS, perinuclear space. B EM quantifications showed that the number of autophagic vacuoles (AV) per neuron increases after both 3 h and 6 h of Tat-BECN1 treatment compared to Tat-sc condition (n ≥ 30, Dunn’s multiple comparisons test; 3 independent experiments). C The percentage of neurons with a focal ballooning of the PNS (not detected in Tat-sc condition) significantly increases between 3 h and 6 h of Tat-BECN1 treatment. (n ≥ 80 neurons per condition; Unpaired t-test; 3 independent experiments). D Tat-BECN1-induced neuronal death is sensitive to cardiac glycosides. The percentage of propidium iodide (PI) positive nuclei is also significantly decreased by digoxin (dig, 50 µM) treatment at 6 h. n ≥ 25 per condition, Dunn’s multiple comparison test; 3 independent experiments). E Representative confocal microscopy images of RFP-GFP-LC3 transfected primary cortical neurons and F corresponding quantifications showing the effect of dig on Tat-BECN1-induced autophagy flux. The number of RFP+-GFP+-LC3-positive dots (yellow), of RFP+-GFP--LC3-positive dots (red) and of the total LC3-positive dots are significantly reduced in the presence dig after 3 h of Tat-BECN1 treatment. Dunn’s multiple comparison test, n ≥ 20 per condition, 3 independent experiments. Scale bar = 10 µm. *p < 0.05, **p < 0.01, ***p < 0.001.
Fig. 4
Fig. 4. Tat-BECN1-induced neuronal autosis is dependent on ATP1a3.
A Representative confocal microscopy images of ATP1a1 and ATP1a3 (in red) in primary cortical neurons (MAP2, in green). Scale bar = 10 µm. B Downregulation of ATP1a1 (with both shRNA atp1a1#1 and shRNA atp1a1#2) has no effect on the percentage of propidium iodide (PI)-positive nuclei induced by 6 h of Tat-BECN1 treatment whereas downregulation of ATP1a3 (with both shRNA atp1a3#1 and shRNA atp1a3#2) significantly reduces the number of PI-positive nuclei (n ≥ 24 per condition; Dunn’s multiple comparisons test; 3 independent experiments). C Representative confocal microscopy images of RFP-GFP-LC3 transfected primary cortical neurons of Tat-sc and Tat-BECN1-treated neurons transduced with shRNAs atp1a1#1, atp1a1#2, atp1a3#1 and atp1a3#2. Scale bar = 10 µm. D Downregulation of ATP1a1 has no significant effect 3 h after Tat-BECN1 treatment on the number of RFP+-GFP+LC3 positive dots (yellow, early autophagosomes), of RFP+-GFP−-LC3 positive dots (red, late autophagosomes) and on the total number of positive dots. In Tat-BECN1 conditions, downregulation of ATP1a3 significantly decreases the number of RFP+-GFP+-LC3 positive dots, the number of RFP+-GFP-LC3 positive dots and the total number of LC3-positive dots (n ≥ 30 per condition, Dunn’s multiple comparison test, 3 independent experiments). Values are mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ns: not significant.
Fig. 5
Fig. 5. Kainate-Hypoxia-induced neuronal death involves ATP1a3-dependent autosis in primary cortical neurons.
A Representative electron micrographs of primary cortical neurons 3 h after CTw or kainate-hypoxia (KaHx) treatment. KaHx-treated primary cortical neuronal cultures showed some dying neurons with morphological characteristics of (iii-iv) phase 1a autosis such as dilatation and fragmentation of endoplasmic reticulum; presence of numerous autophagosomes, autolysosomes and empty vacuoles in the cytosol; (v–vi) phase 1b autosis with restricted swollen perinulear space (PNS) regions (white arrows) containing clumps of cytoplasmic material; and (vii–xii) phase 2 autosis, as shown for different dying neurons, with the presence of a focal ballooning of the (empty) perinuclear space (PNS) associated with a concavity of the nucleus, a swelling of almost all organelles and loss of plasma membrane integrity (black arrows). These morphological features are not observed in CTw-treated neurons (i, ii). N: nucleus, ONM: outer nuclear membrane; INM: inner nuclear membrane; PNS, perinuclear space. B The number of autophagic vacuoles (AV) per neuron increases at both 3 h and 6 h after KaHx treatment compared to CTw condition (n ≥ 30; Dunn’s multiple comparisons test; 3 independent experiments). C The percentage of dying neurons with a focal ballooning of the PNS (not observed in CTw condition) significantly increases between 3 h and 6 h after KaHx treatment (n ≥ 90 neurons per condition; Mann-Whitney test; 3 independent experiments). D Downregulation of ATP1a1 (with both shRNA atp1a1#1 and shRNA atp1a1#2) has no effect on the percentage of propidium iodide (PI)-positive nuclei induced by 6 h of KaHx whereas downregulation of ATP1a3 (with both shRNA atp1a3 #and shRNA atp1a3 #2) significantly reduces the number of PI-positive nuclei (n ≥ 42 per condition; Dunn’s multiple comparisons test; 3 independent experiments). E Representative confocal microscopy images of RFP-GFP-LC3 transfected primary cortical neurons transduced with shRNAs atp1a1#1, atp1a1#2, atp1a3#1 and atp1a3#2 after CTw and KaHx treatment. Scale bar = 10 µm. F Downregulation of ATP1a1 has no significant effect on the RFP-GFP LC3-positive dots pattern 3 h after KaHx treatment. In KaHx conditions, downregulation of ATP1a3 significantly decreases the number of RFP+-GFP+LC3-positive dots (yellow, early autophagosomes), the number of RFP+-GFP-LC3-positive dots (red, late autophagosomes) and the total number of LC3-positive dots (n ≥ 30 per condition, Dunn’s multiple comparison test, 3 independent experiments). Values are mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ns not significant.
Fig. 6
Fig. 6. Increased BECN1-ATP1a3 interaction in in vitro and in vivo rat models of neuronal autosis and in dying neurons of human newborns with hypoxic-ischemic encephalopathy.
A Representative immunoblots and B corresponding quantifications of coimmunoprecipitation of BECLIN1 (BECN1) with ATP1a3 in primary cortical neurons treated with Tat-BECN1 or Tat-sc for 3 h (n = 6, unpaired t-test; 4 independent experiments). C Representative images and corresponding quantifications of proximity ligation assays (PLAs) of BECN1 and ATP1a3 in primary cortical neurons treated with Tat-BECN1 or Tat-sc for 3 h (n ≥ 25, Mann Whitney test; 3 independent experiments; scale bar = 10 µm). D Representative immunoblots and E corresponding quantifications of coimmunoprecipitation of BECN1 with ATP1a3 in primary cortical neurons treated with kainate-hypoxia (KaHx) or control CTw for 3 h (n = 6, unpaired t-test; 3 independent experiments). F Representative images and corresponding quantifications of PLAs of BECN1 and ATP1a3 in primary cortical neurons treated with kainate-hypoxia (KaHx) or control CTw for 3 h (n ≥ 60, Mann Whitney test; 2 independent experiments; scale bar = 10 µm). G Representative immunoblots and H corresponding quantifications of coimmunoprecipitation of BECN1 with ATP1a3 in hippocampus samples of rat pups after 6 h and 24 h of hypoxia-ischemia (HI) and in corresponding sham animals (n = 5 animals/condition, Tukey’s multiple comparison test; 4 independent experiments). I Representative images and corresponding quantifications of PLAs of BECN1 and ATP1a3 in CA3 hippocampus region of rat pups after 6 h and 24 h of hypoxia-ischemia (HI) and in corresponding sham animals (n = 20 images analyzed from 4 animals/condition, Dunn’s multiple comparison test; 3 independent experiments; scale bar = 20 µm). J Representative images and corresponding quantifications of PLAs of BECN1 and ATP1a3 in CA3 hippocampus region of human control (n = 5 cases) and hypoxic-ischemic-encephalopathy (HIE; n = 5 cases) newborns (n ≥ 5 images analyzed / case, Mann Whitney test; scale bar = 20 µm). Values are mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ns: not significant.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. D’Arcy MS. Cell death: a review of the major forms of apoptosis, necrosis and autophagy. Cell Biol Int. 2019;43:582–92. doi: 10.1002/cbin.11137. - DOI - PubMed
    1. Fricker M, Tolkovsky AM, Borutaite V, Coleman M, Brown GC. Neuronal cell death. Physiol Rev. 2018;98:813–80. doi: 10.1152/physrev.00011.2017. - DOI - PMC - PubMed
    1. Parzych KR, Klionsky DJ. An overview of autophagy: morphology, mechanism, and regulation. Antioxid Redox Signal. 2014;20:460–73. doi: 10.1089/ars.2013.5371. - DOI - PMC - PubMed
    1. Soto-Burgos J, Zhuang X, Jiang L, Bassham DC. Dynamics of autophagosome formation. Plant Physiol. 2018;176:219–29. doi: 10.1104/pp.17.01236. - DOI - PMC - PubMed
    1. Ravanan P, Srikumar IF, Talwar P. Autophagy: the spotlight for cellular stress responses. Life Sci. 2017;188:53–67. doi: 10.1016/j.lfs.2017.08.029. - DOI - PubMed

Substances

-