Early exposure to general anesthesia disturbs mitochondrial fission and fusion in the developing rat brain

Abstract

Background: General anesthetics induce apoptotic neurodegeneration in the developing mammalian brain. General anesthesia (GA) also causes significant disturbances in mitochondrial morphogenesis during intense synaptogenesis. Mitochondria are dynamic organelles that undergo remodeling via fusion and fission. The fine balance between these two opposing processes determines mitochondrial morphometric properties, allowing for their regeneration and enabling normal functioning. As mitochondria are exquisitely sensitive to anesthesia-induced damage, we examined how GA affects mitochondrial fusion/fission.

Methods: Seven-day-old rat pups received anesthesia containing a sedative dose of midazolam followed by a combined nitrous oxide and isoflurane anesthesia for 6 h.

Results: GA causes 30% upregulation of reactive oxygen species (n = 3-5 pups/group), accompanied by a 2-fold downregulation of an important scavenging enzyme, superoxide dismutase (n = 6 pups/group). Reactive oxygen species upregulation is associated with impaired mitochondrial fission/fusion balance, leading to excessive mitochondrial fission. The imbalance between fission and fusion is due to acute sequestration of the main fission protein, dynamin-related protein 1, from the cytoplasm to mitochondria, and its oligomerization on the outer mitochondrial membrane. These are necessary steps in the formation of the ring-like structures that are required for mitochondrial fission. The fission is further promoted by GA-induced 40% downregulation of cytosolic mitofusin-2, a protein necessary for maintaining the opposing process, mitochondrial fusion (n = 6 pups/group).

Conclusions: Early exposure to GA causes acute reactive oxygen species upregulation and disturbs the fine balance between mitochondrial fission and fusion, leading to excessive fission and disturbed mitochondrial morphogenesis. These effects may play a causal role in GA-induced developmental neuroapoptosis.

Fig. 1

Anesthesia causes acute ultrastructural changes in mitochondria of pyramidal subicular neurons of 8-day-old rats. (A) Mitochondria in the cytoplasm of subicular pyramidal neurons from sham control animals resemble long tubules with intact inner and outer membranes and numerous cristae tightly packed inside healthy looking matrix. (B) Mitochondria in the cytoplasm of subicular pyramidal neurons from anesthesia- treated animals are numerous. The mitochondria are round, small, and display globular morphology 24 h postanesthesia exposure (on P8). Their matrix is pale and shows the signs of swelling. Although the inner and outer membranes appear somewhat intact, the cristae seem distorted and difficult to discern (C). N = nucleus.

Fig. 2

Anesthesia induces excessive fission of mitochondria in the soma of pyramidal subicular neurons of 8-day-old rats. (A) Mitochondrial density was assessed by counting the number of mitochondrial profiles per unit area (μm²) of cytoplasmic soma in each pyramidal neuron. There are approximately 30% more mitochondrial profiles in anesthesia-treated neurons compared with controls (*P = 0.0179). (B) The summation of mitochondrial areas, presented as a percent of the cytoplasmic area of pyramidal neurons, reveals that mitochondria in the control and experimental neurons occupy approximately the same area of the cytoplasmic soma (P = 0.8067). (C) Frequency distribution analysis by grouping of mitochondrial area indicates that there are significantly more mitochondria smaller than 0.16 μm² (indicated with horizontal bar) in experimental animals compared with controls (P < 0.001). (D) Cumulative frequency analysis (in percentage), designed to take into account the differences in overall mitochondrial number in control vs. experimental neurons, indicates a leftward shift toward smaller mitochondria after anesthesia treatment, with over 50% of mitochondria in the category of lesser than 0.1 μm². In addition, mitochondria smaller than 0.012 μm² were detected in the anesthesia-treated pyramidal neurons, whereas none that small could be detected in the control neurons (n = 4 control and four experimental pups, five neurons from each pup).

Fig. 3

Anesthesia causes acute reactive oxygen species upregulation. Reactive oxygen species were measured in fresh homogenates of subicular and thalamic tissues obtained from P7 rats immediately after 6 h of anesthesia or sham treatment using a kit that detects hydrogen peroxide as described in Methods. We found that the level of reactive oxygen species in anesthesia-treated animals was increased significantly (about 30%) compared to that in sham controls (*P < 0.0357) (n = 3 rat pups in control group; n = 5 rat pups in the experimental group). P7 = postnatal day 7.

Fig. 4

Anesthesia acutely impairs superoxide dismutase (SOD) but not catalase activity. The activities of SOD and catalase were measured in fresh homogenates of subicular and thalamic tissues obtained from P7 rat pups immediately after 6 h of anesthesia or sham treatment and are expressed in units per milligram of protein. (A) We found a significant 2-fold decrease in SOD activity immediately after anesthesia treatment compared to that in sham controls (**P = 0.0011) (n = 6 pups in the control group; n = 6 pups in experimental group). (B) There was no difference in catalase activity between the sham control and experimental groups (P = 0.6631) (n = 6 pups in the control group; n = 6 pups in the experimental group). P7 = postnatal day 7.

Fig. 5

Anesthesia decreases expression of mitofusin-2 (Mfn-2) in the cytosolic fraction. The expression of Mfn-2 protein was estimated from Western blotting in fresh cytosolic and mitochondrial fractions of subicular and thalamic tissues obtained from P7 rats immediately postanesthesia or sham treatment. The protein levels are estimated from Western blotting as percent change from sham controls after normalization to β-actin (cytosolic fraction) or porin (mitochondrial fraction). (A) In the anesthesia- treated group (Treat), Mfn-2 protein expression in the cytosolic fraction was decreased by about 40% compared to that in the sham controls (Cont) (*P = 0.026) (n = 6 pups in the control group; n = 6 pups in the experimental group). (B) In the anesthesia-treated group (Treat), Mfn-2 protein expression in the mitochondrial fraction was approximately the same as that in the experimental group compared to that in the sham controls (Cont) (P = 0.0745) (n = 9 pups in control group; n = 9 pups in experimental group). The molecular mass standards (in kDa) are shown at the right of the representative Western blots (C: cytosolic Mfn-2; D: mitochondrial Mfn-2). (*) Indicates a nonspecific band detected by anti-Mfn2 antibody. P7 = postnatal day 7.

Fig. 6

Anesthesia decreases expression of Drp-1 in the cytosolic fraction and increases Drp-1 in the mitochondrial fraction. The expression of Drp-1 protein was estimated by Western blotting in fresh cytosolic and mitochondrial fractions of subicular and thalamic tissues obtained from P7 rat pups immediately postanesthesia or sham treatment. The protein levels were expressed as percent change from sham controls after normalization to β-actin (cytosolic fraction) or porin (mitochondrial fraction). (A) In the anesthesia-treated group (Treat), Drp-1 protein expression in the cytosolic fraction was significantly decreased compared to sham controls (Cont) (***P < 0.0001) (n = 11 pups in control group; n = 11 pups in experimental group). (B) In the anesthesia-treated group (Treat), Drp-1 protein expression in the mitochondrial fraction was significantly increased compared to that in sham controls (Cont) (***P = 0.0002) (n = 10 pups in the control group; n = 10 pups in the experimental group). The molecular mass standards (in kDa) are shown at the right of the representative Western blots (C = cytosolic Drp-1; D = mitochondrial Drp-1). (*) Indicates alternate splice variant typically recognized by antibody against Drp-1. Drp-1 = dynamin-related protein 1; P7 = postnatal day 7.

Fig. 7

Anesthesia enhances Drp-1 oligomerization in the mitochondria. The samples for Drp-1 oligomer analysis were subjected to a nonreducing gel sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Anesthesia (Treat) increases the protein content of the oligomerized form of Drp-1 in the mitochondrial fraction by about 45% compared to sham controls (Cont) (**P = 0.0037; n = 7 pups in the control group; n = 7 pups in the experimental group). The molecular mass standards (in kDa) are shown at the right of the representative Western blots. Drp-1 = dynamin-related protein 1.

Fig. 8

Proposed pathways that may be responsible for the excessive mitochondrial fission caused by an early exposure to anesthesia. Anesthesia causes downregulation of superoxide dismutase activity accompanied by a lack of compensatory modulation of catalase activity, and these effects are associated with reactive oxygen species upregulation. Elevated reactive oxygen species differentially modulate mitochondrial fission and fusion. They are suggested to induce acute downregulation of Drp-1 protein in the cytoplasm due to its translocation to mitochondria, followed by its oligomerization on the outer mitochondrial membrane, a necessary step in the formation of the ring-like structures required for mitochondrial fission. General anesthesia also causes acute downregulation of mitofusin-2 (Mfn-2), a protein necessary for mitochondrial fusion, thus tipping the fine equilibrium between fission and fusion toward excessive mitochondrial fission. Mitochondria that undergo excessive fission are less functional and more likely to generate excessive amounts of reactive oxygen species, thus further promoting reactive oxygen species upregulation in the setting of downregulated superoxide dismutase activity. In addition, down-regulation of Mfn-2 in the cytoplasm disturbs the redox balance in the neuron, leading to additional reactive oxygen species accumulation. Drp-1 = dynamin-related protein 1.