Neurons Release Injured Mitochondria as "Help-Me" Signaling After Ischemic Stroke

Abstract

Mitochondrial dysfunction has been regarded as one of the major contributors of ischemic neuronal death after stroke. Recently, intercellular mitochondrial transfer between different cell types has been widely studied and suggested as a potential therapeutic approach. However, whether mitochondria are involved in the neuron-glia cross-talk following ischemic stroke and the underlying mechanisms have not been explored yet. In this study, we demonstrated that under physiological condition, neurons release few mitochondria into the extracellular space, and the mitochondrial release increased when subjected to the challenges of acidosis, hydrogen peroxide (H₂O₂), N-methyl-D-aspartate (NMDA), or glutamate. Acidosis reduced the mitochondrial basal respiration and lowered the membrane potential in primary-cultured mouse cortical neurons. These defective mitochondria were prone to be expelled to the extracellular space by the injured neurons, and were engulfed by adjacent astrocytes, leading to increased astrocytic expressions of mitochondrial Rho GTPase 1 (Miro 1) and mitochondrial transcription factor A (TFAM) at mRNA level. In mice subjected to transient focal cerebral ischemia, the number of defective mitochondria in the cerebrospinal fluid increased. Our results suggested that the neuron-derived mitochondria may serve as a "help-me" signaling and mediate the neuron-astrocyte cross-talk following ischemic stroke. Promoting the intercellular mitochondrial transfer by accelerating the neuronal releasing or astrocytic engulfing might be a potential and attractive therapeutic strategy for the treatment of ischemic stroke in the future.

Keywords: ischemic stroke; metabolic stress; mitochondrial biogenesis; mitochondrial release; neuron-glial crosstalk.

FIGURE 1

Extracellular ATP detected in the CM and mdCM of CHO cells and neurons. (A) The CM was collected by spin cell debris down with centrifuging at 2,000 rpm for 10 min, and the mdCM was collected by passing the CM through the 0.2 μm sterile filter. (B) Significant decrease of extracellular ATP concentration in CHO cells was observed after removing mitochondria from the CM of CHO cells. (C) No changes in extracellular ATP concentration were observed after removing mitochondria from the CM of neurons. n = 4. **p < 0.01 vs. CM. CM, culture medium; mdCM, mitochondria-depleted CM; n.s., not significant.

FIGURE 2

Neurons released mitochondria under the acidosis. (A) Schematic diagram describing the experimental procedure. (B) The presence of neuronal mitochondria in culture medium (CM) and mitochondria-depleted culture medium (mdCM) in both groups was analyzed by flow cytometry. (C) The neuronal mitochondria labeled with MitoTracker Green increased in CM treated with a pH 6.5 solution when compared with the pH 7.4 group, but decreased significantly in mdCM in both pH 7.4 and 6.5 groups. (D) The extracellular ATP increased significantly in CM after pH 6.5 treatment, but the ATP concentration decreased dramatically in mdCM after pH 6.5 treatment. n = 4. ***p < 0.001, ****p < 0.0001 vs. pH 7.4 CM; ^###p < 0.001, ^####p < 0.0001 vs. pH 6.5 CM.

FIGURE 3

Neurons released mitochondria under hydrogen peroxide (H₂O₂)-mediated oxidative stress and glutamate- and N-methyl-D-aspartate (NMDA)-induced excitotoxicity. (A) The experimental procedure. (B) The extracellular ATP increased dose-dependently in CM after H₂O₂ treatment. (C,D) The extracellular ATP increased significantly after treatment with NMDA (50 μM) or glutamate (100 μM). n = 4. *p < 0.05, ***p < 0.001 vs. control, ^#p < 0.05 vs. 25 μM H₂O₂.

FIGURE 4

Oxygen consumption rate (OCR) and mitochondrial membrane potential (Δψm; JC-1 red/green ratio) were performed to examine the function of neuronal mitochondria following acidosis. (A) Representative OCR traces ± SD measured in neurons treated with pH 6.5 and 7.4 solutions. (B) Quantification of basal, “leak,” “maximal” state, and ATP production for pH 6.5- and 7.4-treated neurons. (C–E) The JC-1 staining in neurons showed that both JC-1 red and JC-1 green increased after pH 6.5 treatment, but JC-1 red/green ratio decreased significantly. n = 4. *p < 0.05, **p < 0.01 vs. pH 7.4.

FIGURE 5

The mitochondria released from neurons were taken up by astrocytes. (A) Experimental schematic to detect the cross-talk between neurons and astrocytes after treatment with pH 7.4 and 6.5 solutions. The neurons were labeled by MitoTracker Red CMXRos, and the astrocytes were labeled with GFAP. (B) The mitochondria with MitoTracker Red CMXRos were rarely detected within the astrocytes after co-culture with CM from pH 7.4-treated neurons. However, the co-localization of MitoTracker Red CMXRos with GFAP was increased when co-culture with CM from pH 6.5-treated neurons, and the increase was abolished by removing the mitochondria with a 0.2 μm filter.

FIGURE 6

The mRNA expressions of Miro1 and TFAM in astrocytes co-cultured with CM and mdCM from pH 7.4- and 6.5-treated neurons. The mRNA level of Miro1 (A) and TFAM (B) increased significantly in the CM from pH 6.5-treated group but no significant difference in the mdCM group. n = 6. *p < 0.05, ***p < 0.001 vs. pH7.4 CM; ^###p < 0.001 vs. pH 6.5 mdCM. CM, culture medium; mdCM, mitochondria-depleted CM; Miro-1, mitochondrial Rho GTPase 1; TFAM, mitochondrial transcription factor A.

FIGURE 7

The detection of mitochondria and Δψm (JC-1 red/green ratio) in the CSF of MCAO mice. (A) After cerebral ischemia, the MitoTracker Green-labeled mitochondria elevated significantly in CSF at 24-h post-reperfusion. n = 8. *p < 0.05 vs. Sham. (B) The JC-1 staining in CSF showed no significant difference of JC-1 red between the Sham and MCAO groups. (C) JC-1 green increased significantly in the CSF after MCAO when compared with the Sham group. (D) JC-1 red/green ratio decreased significantly in the CSF after MCAO. n = 6. *p < 0.05 vs. Sham. CSF, cerebrospinal fluid; MCAO, middle cerebral artery occlusion; n.s., not significant.

FIGURE 8

Acidosis, oxidative stress, and excitotoxicity are closely associated with mitochondrial dysfunction and are the major culprits that contribute to the ischemic neuronal death after stroke. When neurons are confronting with these harmful challenges, they might release the defective mitochondria to act as a “help-me” signaling, and recruit the adjacent astrocytes for energy support by promoting astrocytic mitochondrial biogenesis. Solid arrow indicates the confirmed results from this study, and dashed arrow indicates the assumption from previously published studies.