Coenzyme Q₁₀ reduces sevoflurane-induced cognitive deficiency in young mice

Abstract

Background

Anaesthesia can induce cognitive deficiency in young rodents and monkeys. Mitochondrial dysfunction contributes to the anaesthesia-induced neurotoxicity and neurobehavioural deficits. We therefore assessed the effects of the mitochondrial energy enhancer coenzyme Q₁₀ (CoQ₁₀) on anaesthesia-induced cognitive deficiency in young mice to investigate the role of mitochondrial dysfunction.

Methods

Young mice (n=134) were randomly assigned into the following four groups: control plus corn oil vehicle (60% oxygen); 3% sevoflurane [2 h daily on postnatal day (P) 6, 7, and 8] plus vehicle; CoQ₁₀ (50 mg kg⁻¹) plus vehicle; or 3% sevoflurane plus CoQ₁₀ plus vehicle. We determined cognitive function using the Morris water maze at P31-P37. We quantified brain postsynaptic density protein-95, the presynaptic marker synaptophysin, adenosine triphosphate, reactive oxygen species, and mitochondrial membrane potential at P8 and P37.

Results

Coenzyme Q₁₀ reduced sevoflurane-induced cognitive deficiency in young mice (F=0.90, P=0.49, n=10–16) and attenuated sevoflurane-induced reductions in postsynaptic density protein-95 (F=10.56, P<0.01, n=6), synaptophysin (F=8.44, P=0.01, n=6), adenosine triphosphate (F=4.34, P=0.05, n=9), and mitochondrial membrane potential (F=11.43, P<0.01, n=6), but not sevoflurane-induced increases in reactive oxygen species (F=1.17, P=0.20, n=6), in brain.

Conclusions

These data suggest that CoQ₁₀ reduces sevoflurane-induced cognitive deficiency by mitigating sevoflurane-induced mitochondrial dysfunction, the reduction in adenosine triphosphate, and synaptic dysfunction. Coenzyme Q₁₀ could provide an approach to reduce the neurotoxicity of anaesthesia in the developing brain.

Editor's key points.

• •Mitochondria have been implicated as targets for anaesthetic neurotoxicity and cognitive deficits in neonatal rodents exposed to anaesthesia.
• •The endogenous antioxidant coenzyme Q₁₀ reduced sevoflurane anaesthesia-induced cognitive deficits in neonatal mice and attenuated reductions in synaptic markers, adenosine triphosphate, and mitochondrial membrane potential.
• •Coenzyme Q₁₀ protects against sevoflurane-induced mitochondrial dysfunction and energy deficits leading to synaptic damage and cognitive dysfunction in neonatal mice.

Millions of children require surgical procedures and anaesthesia every year around the world. Children who undergo anaesthesia and surgery may have increased risk of cognitive deficiency and changes in brain structure (e.g. reduced grey matter density in both occipital cortex and cerebellum,1, 2, 3 reviewed by Sun⁴ and Vutskits and Xie⁵). Wilder and colleagues¹ reported that children <4 yr old who received multiple (e.g. three times), but not single, episodes of anaesthesia and surgery had increased risk of learning disability before the age of 15 yr. Recent prospective epidemiological studies show that single and short-duration anaesthesia (e.g. sevoflurane) and surgery (e.g. hernia repair) are not associated with cognitive deficiency in children.^{6 7} However, several other studies have shown an association between cognitive deficiency and surgery under anaesthesia in children.^{2 3} 8, 9, 10, 11

Anaesthesia can induce cognitive deficiency and neurotoxicity in young rodents12, 13, 14, 15, 16 and monkeys.17, 18, 19, 20, 21 Mitochondrial dysfunction might contribute to this anaesthetic neurotoxicity. Anaesthesia with midazolam, nitrous oxide, and isoflurane enlarged mitochondrial size, impaired structural integrity of mitochondria, and increased complex IV activity.²² The same combination of anaesthetics increased brain concentrations of reactive oxygen species (ROS) and impaired the balance between mitochondrial fission and fusion, which caused excessive fission and impaired mitochondrial morphogenesis.²³ The NADPH oxidase inhibitor apocynin attenuated sevoflurane-induced cognitive impairment and the increase in brain concentrations of superoxide and NADPH oxidase subunit p22phox.²⁴ Multiple exposures to sevoflurane anaesthesia in young rats caused mitochondrial toxicity and reduced the number of synapses in rat hippocampus.²⁵ Moreover, treatment with the ROS scavenger EUK-134 and restoration of mitochondrial integrity by R(+) pramipexole inhibited the anaesthesia-induced mitochondrial dysfunction and cognitive impairment in rats.²⁶

Our previous experiments showed that anaesthesia can produce mitochondrial dysfunction and suggested the involvement of energy deficiency in anaesthetic neurotoxicity.²⁷ Mitochondrial dysfunction induced by isoflurane contributes to isoflurane-induced apoptosis,28, 29 which might accelerate the neuropathogenesis of Alzheimer's disease (reviewed by Vutskits and Xie⁵). Moreover, mitochondrial dysfunction can impair synapse formation.³⁰ A recent study showed that multiple exposures to sevoflurane in monkeys can induce synaptic loss and mitochondrial dysfunction, and there was a positive association between mean presynaptic mitochondrial density and mean synaptic density after sevoflurane anaesthesia.²⁵ These findings suggest a connection between synaptic loss and mitochondrial dysfunction. However, it remains largely unknown how anaesthesia can induce mitochondrial dysfunction leading to cognitive impairment (e.g. the duration and timing of anaesthesia). The impact of anaesthesia on synaptogenesis or synapse maintenance in developing brain and potential treatments remain to be determined.

The goal of the present proof-of-concept research was to determine whether an energy enhancer could alleviate anaesthesia-induced cognitive deficiency and reduce the anaesthesia-induced reduction in brain concentrations of synaptic markers, adenosine triphosphate (ATP), and mitochondrial function indicators [ROS and mitochondrial membrane potential (MMP)] in young mice. Coenzyme Q₁₀ (CoQ₁₀; ubiquinone) is an endogenous lipid-soluble antioxidant found mainly in the mitochondrial inner membrane.³¹ Coenzyme Q₁₀ is a significant cofactor in the electron transport chain of mitochondria and has neuroprotective, energy-converting, anti-inflammatory, and antioxidant effects.32, 33, 34, 35 Specifically, CoQ₁₀ can improve bioenergetic functions of mitochondria, enhance brain energy levels (e.g. increase ATP), and improve brain function in rodents.^{36 37} Finally, CoQ₁₀ reduces ROS and mitochondria dysfunction36, 37, 38 and reduces amyloid-associated pathology in Alzheimer's disease transgenic mice.39, 40 However, the interaction of CoQ₁₀ and anaesthesia on cognitive function in young mice has not been determined. We therefore assessed whether CoQ₁₀ mitigates sevoflurane-induced cognitive deficiency and probed the underlying mechanisms. This included measurement of the effects of sevoflurane on the concentrations of synaptic markers, specifically synaptophysin,^{41 42} a presynaptic marker, and postsynaptic density protein-95 (PSD-95),43, 44, 45, 46, 47, 48 an excitatory postsynaptic marker. Our previous studies have shown that sevoflurane anaesthesia reduces PSD-95 in the young mouse hippocampus.¹⁶ Sevoflurane anaesthesia in pregnant mice also reduces amounts of both PSD-95 and synaptophysin in fetal mouse brain.⁴⁹ We tested the hypothesis that CoQ₁₀ reduces sevoflurane-induced cognitive deficiency and alleviates sevoflurane-induced reductions in PSD-95, synaptophysin, ROS, MMP, and ATP in young mouse hippocampus and cortex.

Methods

Anaesthesia and treatment of mice

We completed all tests according to the guidelines and regulations of National Institutes of Health (NIH). The Standing Committee on the Use of Animals in Research and Teaching at Massachusetts General Hospital approved the study, including the care of animals (protocol number: 2006N000219, Boston, MA, USA). We minimized the number of mice used in the tests. We used both female and male mice (C57BL/6J; Jackson Laboratories, Bar Harbor, ME, USA) but did not assess potential sex differences in the present studies.

A total of 134 young mice were assigned randomly into the following four groups: control plus corn oil; control plus CoQ₁₀; sevoflurane plus corn oil; and sevoflurane plus CoQ₁₀. The mice were placed in an anaesthesia chamber to receive the anaesthesia or control using 3% sevoflurane plus 60% oxygen (balanced with nitrogen) for 2 h daily for 3 days from postanatal day (P) 6 to P8 as performed in our previous studies.^{15 16} The sevoflurane concentration of 3% is a clinically relevant concentration, and anaesthesia for 2 h daily for 3 days from P6 to P8 conceptually mimics multiple exposures of anaesthesia in young patients. The control condition was oxygen (60% oxygen, balanced with nitrogen) with an equal rate of flow in a chamber similar to the anaesthesia chamber.^{15 16} Mice in the control conditions were separated from the dams and given the control gas (60% oxygen) exposures (Fig. 1). Specifically, 24 mice were decapitated at the end of the anaesthesia at P8 for harvest of the hippocampus, which was used in the western blot studies. Thirty-six mice were decapitated at the end of the anaesthesia at P8 for harvest of the cortex for ATP measurement. Twenty-four mice were decapitated at the end of anaesthesia at P8 for harvest of the hippocampus and cortex for measurement of ROS (hippocampus) and MMP (mitochondria from cortex). Fifty mice were used for Morris water maze (MWM) from P31 to P37 and western blot studies at P37. The anaesthesia chamber size was 20 cm (length) × 20 cm (width) × 7 cm (height). The induction fresh gas flow rate was 2 litres min⁻¹ for 3 min (for the purpose of induction) and then 1 litre min⁻¹ (for maintenance). The anaesthetic and oxygen were continuously sampled from the distal portion of the cylindrical chamber using a gas analyser (Dash 4000; GE Healthcare, Milwaukee, WI, USA) during anaesthesia. We monitored and controlled the anaesthesia chamber temperature via a feedback-based system (DC Temperature Control System; World Precision Instruments, Inc., Sarasota, FL, USA) to control rectal temperature at 37 °C (0.5 °C) by placing a warming pad under the chamber. Previous studies^{14 50} have shown that anaesthesia with 3% sevoflurane did not significantly modify pH values, oxygen partial pressure, or carbon dioxide partial pressure in young mice. We therefore did not measure blood gases in the present studies. We assessed food and water consumption by the quantity of food and water used and by variations in mouse weight during the study. In the intervention studies, we gave CoQ₁₀ (50 mg kg⁻¹, dissolved in corn oil at a concentration of 1.5 µg µl⁻¹ (Sigma-Aldrich, Inc., St Louis, MO, USA)⁵¹ by intraperitoneal (i.p.) administration 30 min before sevoflurane anaesthesia on P6, P7, and P8. Mice in the control group received 100 µl corn oil i.p.

Fig 1.

Diagram of experimental design. Young mice (n=134) received anaesthesia with 3% sevoflurane for 2 h or control on 3 days at postnatal day (P) 6, P7, and P8. We harvested brain tissue (hippocampus and cortex) at the end of anaesthesia on P8 for western blotting (n=24), measurement of adenosine triphosphate (ATP; n=36), and measurement of reactive oxygen species (ROS) and mitochondrial membrane potential (MMP; n=24). Different groups of mice (n=50) were used for assessment of learning and memory function using the Morris water maze from P31 to P37 after the birth. We harvested hippocampus at the end of the Morris water maze assessments (P37) for western blotting (n=50).

Morris water maze experiments

We performed MWM experiments using published protocols.¹⁵ We tested P31 mice in the MWM for 7 days (from P31 to P37) with four tests daily. We used a video recording device to track mice swimming in the pool. Both escape latency (time to reach the platform) and platform crossing (times the mouse moved across the original area of the removed platform) were recorded on P37 to assess spatial learning and memory capacities. We maintained mouse body temperature using a heating device as described^{15 16 52} by placing each mouse under a heat lamp for 5 min in a holding cage to dry the mouse. We then returned the mouse to its home cage.

Harvest of brain tissue and quantification of proteins

Mice were allowed to totally recover from anaesthesia. We killed mice by decapitation one hour after the end of sevoflurane anaesthesia on P8 or P37, and harvested the hippocampus and cerebral cortex for analysis of synaptic proteins, ROS, MMP and ATP. We homogenized the harvested hippocampus or cortex on ice in immunoprecipitation buffer (Tris-HCl: 10 mM, pH 7.4; NaCl: 150 mM; EDTA: 2 mM, Nonidet P-40: 0.5%) plus protease inhibitors (aprotinin: 1 μg ml⁻¹; leupeptin: 1 μg ml⁻¹, pepstatin A: 1 μg ml⁻¹). Lysates were centrifuged for 10 min at 14800g. Total protein was quantified using a bicinchoninic acid protein assay kit (Pierce, Iselin, NJ, USA).²⁷

Western blot analysis

We used PSD-95 antibody (Cell Signaling, Danvers, MA, USA; #2507) at 1:1000 dilution, synaptophysin antibody (Cell Signaling; #4329) at 1:1000 dilution, and a β-actin antibody (Sigma, St Louis, MO, USA; #A5441) at 1:5000 dilution for quantification by western blot analysis as described by Zhang and colleagues.²⁷ We analysed the signal intensity using the Quantity One image analysis program (Bio-Rad, Hercules, CA, USA). We used β-actin concentrations to standardize amounts of protein (e.g. calculating the proportion of PSD-95 in relationship to the quantity of β-actin) and limit disparities in the quantity of protein loaded. We expressed the protein concentrations as a percentage in relationship to control mice.

Measurement of ATP

We used another cohort of mice for ATP measurement performed using the ATP Colorimetric/Fluorometric Assay Kit (BioVision, Milpitas, CA, USA).53, 54, 55, 56 We homogenized cerebral cortex on ice using 100 µl ATP assay buffer. We then we took 80 µl of sample and 20 µl ice-cold perchloric acid and put them on ice for 5 min with vortexing. We centrifuged the samples for 2 min at 13 000g. The supernatant (76 µl) was transferred to a new tube, and 4 µl ice-cold neutralization solution was added, mixed to neutralize the sample, placed on ice, and opened to the air for 5 min. We then centrifuged the tube at 13 000g for 2 min for the bicinchoninic acid assay. We added sample and ATP assay buffer (1 mM) to 50 µl per well in a 96-well plate, and added reaction mix of 50 µl to each well containing the ATP standard or test samples. The reaction mix included 44 µl ATP assay buffer, 2 µl ATP probe, 2 µl ATP converter, and 2 µl developer, and samples were incubated for 30 min at room temperature (avoiding light). We measured the absorbance (optical density at 570 nm) in a microplate reader to calculate the concentration of ATP.

Measurement of ROS

An OxiSelect In Vitro ROS/RNS Assay Kit (Cell Biolabs, San Diego, CA, USA) was used to measure ROS in vivo according to protocols provided by company and used in previous studies.57, 58 Specifically, the harvested brain tissues (hippocampus) were homogenized on ice using 1% Triton-100 in phosphate-buffered saline plus protease inhibitors (1 μg ml⁻¹ aprotinin, 1 μg ml⁻¹ leupeptin, and 1 μg ml⁻¹ pepstatin A). Lysates were collected, centrifuged at 10 300g for 5 min, and quantified for total protein by bicinchoninic acid protein assay. We added 50 μl of each sample (triplicate) to a black 96-well plate and 50 μl of catalyst solution, and then incubated the mixture for 5 min at room temperature. We then added 100 μl of dichlorodihydrofluorescin (DCFH) solution included in the kit to the samples, and incubated the mixture at room temperature for 20 min away from light. Fluorescence was read with a fluorometric plate reader at excitation 480 nm, emission 530 nm.

Isolation of mitochondria and determination of mitochondrial membrane potential

We used a mitochondrial isolation kit (Thermo Fisher Scientific, Waltham, MA, USA) to isolate mitochondria from brain to detect MMP as described.^{28 59} Specifically, cerebral cortex was placed in a 2 ml microcentrifuge tube and cut into small pieces in 800 μl of phosphate-buffered saline. Tubes were centrifuged at 1000g for 3 min, and the supernatant was removed and discarded. Then we added 800 μl of mitochondrial isolation reagent A into each of the tubes, and the tubes were vortexed for 5 s and incubated on ice for 5 min. We added 10 μl of mitochondrial isolation reagent B into each of the tubes, and each tube was vortexed for 5 s and incubated on ice for 5 min before adding 800 μl of mitochondrial isolation reagent C. Tubes were centrifuged at 700g for 10 min at 4 °C, and the supernatant was transferred to a new 1.5 ml tube and centrifuged at 3000g for 15 min. We collected the pellet, which contained isolated mitochondria. Protein concentrations in the mitochondrial pellets were measured using the bicinchoninic acid protein assay. Freshly isolated mitochondria were used immediately for determination of MMP.

We used a JC-1 (5,50,6,60-tetrachloro-1,10,3,30 tetraethylbenzimidazolylcarbocyanine iodide) mitochondrial membrane potential detection kit (Biotium, Hayword, CA, USA) to determine MMP according to the manufacturer's protocol as described by Baumber and colleagues.⁶⁰ The MMP was calculated by detection of the JC-1 fluorescence ratio. Specifically, isolated mitochondria were resuspended in 0.5 ml 1× JC-1 reagent, incubated at 37°C in a 5% CO₂ incubator for 15 min, centrifuged for 5 min at 400g, and the pellet was resuspended in 2 ml 1× assay buffer followed by centrifugation. We resuspended the pellet in 0.3 ml 1× assay buffer and measured red (excitation 550 nm, emission 600 nm) and green (excitation 485 nm, emission 535 nm) fluorescence using a fluorescence plate reader. The ratio of red to green fluorescence is decreased in dead samples and in samples undergoing apoptosis compared with healthy samples.

Statistics

We expressed data obtained from biochemistry studies and MWM escape latency as the means (sd). Numbers of platform crossing of MWM were not normally distributed, and they are expressed as medians with the interquartile range. The number of samples was 10–16 per group in behavioural studies and six to nine per group in biochemistry studies. These numbers were selected according to the results obtained in our previous studies. We used two-way anova (repeated measurements) to evaluate the difference of escape latency of mice in the anaesthesia cohort compared with the control cohort in the MWM. We used post hoc analysis to contrast the change in escape latency for each day during the MWM test, and cut-off α was Bonferroni adjusted. We used the Mann–Whitney U-test to compare the platform-crossing times of mice in the anaesthesia cohort with mice in the control cohort. We used two-way anova to evaluate the interaction of group (control vs anaesthesia) and treatment (CoQ₁₀ vs corn oil) on the amounts of PSD-95, synaptophysin, ATP, ROS, and MMP. Two-tailed hypothesis testing was used. Statistical significance was defined as P<0.05. We used Prism 6 (GraphPad, La Jolla, CA, USA) to evaluate all data.

Results

Coenzyme Q₁₀ treatment reduced sevoflurane-induced cognitive deficiency

We determined whether CoQ₁₀, a mitochondrial energy enhancer,^{36 37} could reduce the anaesthesia-induced cognitive deficiency in young mice. In mice pretreated with corn oil (vehicle), anaesthesia with sevoflurane (3% sevoflurane 2 h daily for three days) on P6, P7, and P8 induced cognitive deficiency in the MWM test from P31 to P37. There was a borderline significant interaction (F=2.17, P=0.05, two-way anova) between the group (control vs anaesthesia) and time (P31–P37). Sevoflurane anaesthesia increased the time (escape latency) for mice to locate the platform in the MWM pool compared with the control conditions (Fig. 2A). Post hoc testing (Bonferroni) showed that the anaesthesia mice took longer (escape latency) to locate the platform compared with control mice at P35 (P<0.01) and P37 (P<0.05; Fig. 2A). Sevoflurane anaesthesia also reduced platform-crossing times compared with control conditions (Fig. 2B; P=0.01, Mann–Whitney U-test).

Fig 2.

Coenzyme Q₁₀ (CoQ₁₀) attenuated sevoflurane-induced cognitive deficiency. (A) There was a non-significant interaction between group (control vs sevoflurane) and time [postnatal day (P) 31–P37] in the escape latency of the Morris water maze (MWM) test with pretreatment of corn oil (CoQ₁₀ vehicle). (B) There was a significant difference in platform-crossing times between control mice and and mice exposed to sevoflurane anaesthesia. (C) There was not a significant interaction between group (control vs sevoflurane) and time (P31–P37) in MWM escape latency for pretreatment with CoQ₁₀. (D) There was not a significant difference in platform-crossing times between control mice and the mice exposed to sevoflurane anaesthesia and pretreatment with CoQ₁₀. n=10–16 in each group.

In mice pretreated with CoQ₁₀, sevoflurane anaesthesia did not induce cognitive deficiency tested with the MWM. No significant interaction (F=0.90, P=0.49) between group (control and anaesthesia) and time (P31–P37) was observed, and sevoflurane anaesthesia did not significantly increase the time (escape latency) for mice to locate the platform (Fig. 2C). Sevoflurane anaesthesia did not decrease platform-crossing times compared the control conditions (Fig. 2D; P=0.78, Mann–Whitney U-test).

Coenzyme Q₁₀ reduced sevoflurane-induced reductions in PSD-95 and synaptophysin in hippocampus

Given that CoQ₁₀ alleviated sevoflurane-induced cognitive deficiency, we inquired whether CoQ₁₀ could reduce sevoflurane-induced changes in synaptic markers. Synaptophysin^{41 42} and PSD-9543, 44, 45, 46, 47, 48 are pre- and postsynaptic markers, respectively. Sevoflurane has been reported to reduce amounts of PSD-95 in young mouse hippocampus^{16 52} and to reduce amounts of PSD-95 and synaptophysin in fetal mouse brain.⁴⁹ We investigated the effects of CoQ₁₀ on the amounts of PSD-95 and synaptophysin in hippocampus of the young mice after sevoflurane anaesthesia. Immunoblotting analysis showed that sevoflurane anaesthesia (Fig. 3A) reduced amounts of PSD-95 in hippocampus compared with control conditions at P8. Coenzyme Q₁₀ (Fig. 3A) itself did not significantly modify amounts of PSD-95 compared with vehicle. However, CoQ₁₀ attenuated sevoflurane-induced reductions in amounts of PSD-95 (Fig. 3A) at P8. Western blot quantification showed that sevoflurane (Fig. 3B) reduced amounts of PSD-95 compared with control conditions, and CoQ₁₀ attenuated the reduction in PSD-95 induced by sevoflurane anaesthesia at P8 (F=10.56, P<0.01, two-way anova).

Fig 3.

Coenzyme Q₁₀ (CoQ₁₀) attenuated the sevoflurane-induced reduction in postsynaptic density protein-95 (PSD-95) in hippocampus. (A) Anaesthesia with 3% sevoflurane for 2 h daily for 3 days decreased PSD-95 concentrations in the hippocampus compared with control conditions at postnatal day (P) 8. Coenzyme Q₁₀ attenuated the sevoflurane-induced reduction in PSD-95 in hippocampus compared with corn oil at P8. (B) Quantification of western blot data showed that sevoflurane anaesthesia reduced PSD-95 concentrations in hippocampus compared with control conditions. Coenzyme Q₁₀ attenuated the sevoflurane-induced reduction in PSD-95 in hippocampus compared with corn oil. (C) Western blotting showed that anaesthesia with 3% sevoflurane for 2 h daily for 3 days at P6, P7, and P8 decreased PSD-95 concentrations in hippocampus compared with control at P37. Coenzyme Q₁₀ attenuated sevoflurane-induced reduction in PSD-95 in hippocampus compared with corn oil at P37. (D) Quantification of western blot data showed that the sevoflurane anaesthesia reduced PSD-95 in hippocampus compared with control conditions at P37. Coenzyme Q₁₀ attenuated the sevoflurane-induced reduction in PSD-95 in hippocampus compared with corn oil at P37. n=6 in each group.

We also found that sevoflurane anaesthesia (Fig. 3C and D) decreased PSD-95 in hippocampus compared with control conditions at P37. Coenzyme Q₁₀ (Fig. 3C and D) itself did not significantly modify amounts of PSD-95 compared with vehicle. However, CoQ₁₀ attenuated the sevoflurane-induced reduction in amounts of PSD-95 at P37 (F=10.88, P<0.01, two-way anova; Fig. 3C and D). Sevoflurane anaesthesia also reduced synaptophysin concentrations in hippocampus compared with control conditions (Fig. 4A and B) at P8. Coenzyme Q₁₀ attenuated sevoflurane-induced reduction in amounts of synaptophysin at P8 (Fig. 4A and B; F=8.44, P=0.01, two-way anova). Finally, sevoflurane anaesthesia decreased synaptophysin concentrations in hippocampus compared with control conditions at P37 (Fig. 4C and D), and CoQ₁₀ attenuated sevoflurane-induced reductions in amounts of synaptophysin at P37 (F=24.78, P<0.01, two-way anova; Fig. 4C and D). Collectively, these data show that CoQ₁₀ attenuates both the cognitive deficiency in young mice from P31 to P37 and the reduction in amounts of PSD-95 and synaptophysin (synaptic markers) in hippocampus at P8 and P37 induced by sevoflurane anaesthesia.

Fig 4.

Coenzyme Q₁₀ (CoQ₁₀) attenuated the sevoflurane-induced reduction in synaptophysin in the hippocampus. (A) Western blotting showed that anaesthesia with 3% sevoflurane 2 h daily for 3 days decreased synaptophysin concentrations in hippocampus compared with control conditions at postnatal day (P) 8. Coenzyme Q₁₀ attenuated the sevoflurane-induced decrease in synaptophysin in hippocampus compared with corn oil at P8. (B) Quantification of western blot data showed that sevoflurane anaesthesia reduced synaptophysin concentrations in hippocampus compared with control conditions at P8. Coenzyme Q₁₀ attenuated the sevoflurane-induced reduction in synaptophysin in hippocampus compared with corn oil at P8. (C) Western blotting showed that anaesthesia with 3% sevoflurane for 2 h daily for 3 days at P6, P7, and P8 decreased synaptophysin in hippocampus compared with control conditions at P37. Coenzyme Q₁₀ attenuated the sevoflurane-induced decrease in synaptophysin in hippocampus compared with corn oil at P37. (D) Quantification of western blot data showed that sevoflurane anaesthesia reduced synaptophysin in hippocampus compared with control conditions at P37. Coenzyme Q₁₀ attenuated the sevoflurane-induced reduction in synaptophysin in hippocampus compared with corn oil at P37. n=6 in each group.

Coenzyme Q₁₀ reduced sevoflurane-induced reductions in ATP in cerebral cortex

Next, we determined the effects of CoQ₁₀ on sevoflurane-induced changes in ATP mouse cerebral cortex. Two-way anova presented a borderline significant interaction of group (control vs anaesthesia) and treatment (corn oil vs CoQ₁₀) on amounts of ATP (F=4.343, P=0.05; Fig. 5). Specifically, sevoflurane anaesthesia decreased amounts of ATP in cortex compared with control conditions (Fig. 5). Coenzyme Q₁₀ attenuated the sevoflurane-induced reductions in amounts of ATP (Fig. 5). Collectively, these data further suggest that CoQ₁₀ reduced the cognitive deficiency and attenuated reductions in PSD-95, synaptophysin, and ATP induced by sevoflurane anaesthesia.

Fig 5.

Coenzyme Q₁₀ (CoQ₁₀) attenuated the sevoflurane-induced reduction in adenosine triphosphate (ATP) in cerebral cortex. Anaesthesia with 3% sevoflurane for 2 h daily for 3 days [postnatal day (P) 6, P7, and P8] decreased amounts of ATP in cortex compared with control conditions at P8. Coenzyme Q₁₀ attenuated the sevoflurane-induced reduction in ATP in cortex compared with corn oil at P8. n=9 in each group.

Coenzyme Q₁₀ did not reduce sevoflurane-induced increases in ROS in hippocampus

Given the findings that CoQ₁₀ attenuated the sevoflurane-induced reduction in amounts of ATP, we asked whether CoQ₁₀ could reduce sevoflurane-induced changes in mitochondrial function. Sevoflurane anaesthesia increased ROS compared with control conditions (Fig. 6) in hippocampus at P8. However, CoQ₁₀ did not attenuate the sevoflurane-induced increase in the amount of ROS (F=1.72, P=0.20, two-way anova; Fig. 6).

Fig 6.

Coenzyme Q₁₀ (CoQ₁₀) did not attenuate the sevoflurane-induced increase in reactive oxygen species (ROS) in hippocampus. Anaesthesia with 3% sevoflurane for 2 h daily for 3 days [postnatal day (P) 6, P7, and P8] increased amounts of ROS in hippocampus compared with the control conditions at P8. Coenzyme Q₁₀ did not attenuate the sevoflurane-induced increase in ROS in hippocampus compared with corn oil. n=6 in each group.

Coenzyme Q₁₀ reduced sevoflurane-induced reduction in MMP in cerebral cortex

Finally, we found that sevoflurane anaesthesia decreased the MMP in mitochondria from the cerebral cortex of the mice at P8 compared with control mice (Fig. 7). Moreover, two-way anova showed a significant interaction between group (control vs sevoflurane) and treatment (corn oil vs CoQ₁₀), and CoQ₁₀ attenuated sevoflurane-induced reductions in MMP concentrations in cortex at P8 (F=11.43, P<0.01; Fig. 7). Collectively, these data suggest that sevoflurane anaesthesia decreases brain ATP concentrations in young mice by impairing mitochondrial function, and CoQ₁₀ can mitigate the sevoflurane-induced reduction in ATP by attenuating the sevoflurane-induced reduction in brain MMP, but not the sevoflurane-induced increase in brain ROS at P8.

Fig 7.

Coenzyme Q₁₀ (CoQ₁₀) attenuated the sevoflurane-induced reduction in mitochondrial membrane potential (MMP) in cerebral cortex. Anaesthesia with 3% sevoflurane for 2 h daily for 3 days [postnatal day (P) 6, P7, and P8] decreased MMP in cortex compared with control conditions at P8. Coenzyme Q₁₀ attenuated the sevoflurane-induced reduction in MMP in cortex compared with corn oil. n=6 in each group.

Discussion

In this proof-of-concept study, we assessed whether CoQ₁₀ could lessen the sevoflurane-induced cognitive deficiency in young mice and inhibit sevoflurane-induced changes in synaptophysin, PSD-95, ATP, ROS, and MMP in hippocampus and cerebral cortex. We found that sevoflurane anaesthesia caused cognitive deficiency in young mice, consistent with our previous studies.^{15 16} However, in young mice pretreated with CoQ₁₀, sevoflurane anaesthesia did not induce cognitive deficiency. Given that CoQ₁₀ is a brain energy enhancer,^{36 37} these data suggest that brain energy deficits could contribute, at least in part, to the cognitive deficiency induced by sevoflurane anaesthesia, and CoQ₁₀ could be used to prevent or treat the anaesthesia-induced cognitive deficiency in young mice.

Coenzyme Q₁₀ also attenuated sevoflurane-induced reductions in the amounts of PSD-95 and synaptophysin, which are postsynaptic^{41 42} and presynaptic markers,43, 44, 45, 46, 47, 48 respectively. These data suggest that the underlying mechanisms by which CoQ₁₀ reduces sevoflurane-induced cognitive deficiency include rescue of sevoflurane-induced synaptic dysfunction in the hippocampus.

Coenzyme Q₁₀ also attenuated sevoflurane-anaesthesia-induced decreases in ATP concentrations in cerebral cortex. Collectively, these results that CoQ₁₀ reduced sevoflurane-induced cognitive deficiency and reductions in synaptic markers and energy levels in young mouse brain support the hypothesis that brain energy deficits contribute to the anaesthesia-induced cognitive deficiency. Pending further investigation, these findings demonstrate the potential underlying mechanisms of anaesthesia-induced neurobehavioural deficits and neurotoxicity in the developing brain and suggest that energy enhancers, including CoQ₁₀ and vitamin K₂,⁶¹ might be used to mitigate potential postoperative cognitive decline or dysfunction in children.

Coenzyme Q₁₀ has been reported to augment brain energy levels and enhance cognitive function in rodents. For example, CoQ₁₀ was able to reduce the brain energy deficit and cognitive dysfunction trigged by intracerebroventricular streptozotocin in rats.⁶² Coenzyme Q₁₀ improved cognitive function in Alzheimer's disease transgenic mice,⁶³ and age-associated deficiency⁶⁴ and cognitive deficiency⁶⁵ in mice. Finally, CoQ₁₀ reduced β-amyloid-induced cognitive deficiency in mice.⁶⁶ We show, for the first time, that CoQ₁₀ can reduce anaesthesia-induced cognitive deficiency and rescue anaesthesia-induced brain energy deficits in mice. We postulated that anaesthesia might cause neurotoxicity and neurobehavioural deficits in young mice by producing brain energy deficits. Further investigations are warranted to test this hypothesis.

Activity-driven ATP synthesis is required for synaptic function, and a reduction in activity-stimulated ATP synthesis leads to deficiency of presynaptic function.⁶⁷ We found CoQ₁₀ reduced sevoflurane-induced reductions in both synaptic markers and the concentration of ATP in brain. Thus, it is conceivable that sevoflurane anaesthesia reduces brain energy levels, which then causes synaptic dysfunction, leading to cognitive deficiency.

Moreover, we found that sevoflurane anaesthesia increased ROS concentrations and decreased MMP, but CoQ₁₀ rescued only the sevoflurane-induced reduction in MMP. These data suggest that CoQ₁₀ could rescue the sevoflurane-induced reduction in brain amounts of ATP by attenuating sevoflurane-induced mitochondrial dysfunction. Interestingly, CoQ₁₀ did not rescue sevoflurane-induced increases in brain ROS. Taken together, these results suggest that CoQ₁₀ prevents sevoflurane-induced damage of the electron transport chain or a later stage, but not sevoflurane-induced ROS accumulation or an earlier stage, of mitochondrial dysfunction. Our findings that CoQ₁₀ attenuated sevoflurane-induced cognitive impairment in young mice and changes in PSD-95, synaptophysin, MMP, and ATP indicate that i.p. CoQ₁₀ was able to reach the brain. However, it is unknown whether CoQ₁₀ itself or a metabolite(s) mitigated the sevoflurane-induced cognitive impairment and changes in ATP, PSD-95, synaptophysin, and MMP.

Several studies have demonstrated that anaesthetics can cause mitochondrial dysfunction, including enlargement of the size of mitochondria, impairment of structural integrity of mitochondria, increased complex IV activity, increased amounts of ROS, impaired balance between mitochondrial fission and fusion, and reductions in the fraction of presynaptic terminals containing mitochondria in brain tissues of rodents.^{22 23 25} Moreover, the NADPH oxidase inhibitor apocynin, ROS scavenger EUK-134, and R(+) pramipexole (which can recover mitochondrial integrity) can mitigate anaesthesia-induced mitochondrial dysfunction and cognitive impairment.^{24 26} Our findings that CoQ₁₀ ameliorated sevoflurane-induced cognitive impairment in young mice and reductions in brain concentrations of synaptic markers, ATP, and MMP suggest that CoQ₁₀ prevents or treats anaesthesia-induced cognitive impairment by rescuing mitochondrial dysfunction in the mice. These findings also provide an opportunity to design new interventions to prevent or treat possible anaesthesia-induced cognitive deficiency in children.

Multiple exposures to sevoflurane anaesthesia in young rats caused synaptic loss in rat hippocampus,²⁵ and isoflurane anaesthesia in young rats caused a reduction in synaptic density in the hippocampus.⁶⁸ Anaesthesia with midazolam, nitrous oxide, and isoflurane reduced synapse volumetric densities in the hippocampus of young rats,⁶⁹ and sevoflurane decreased the synaptic marker PSD-95 in the hippocampus of young mice.^{16 52} However, the upstream mechanism by which anaesthetic reduces concentrations of synaptic markers and synaptic function remains largely to be investigated.

Our studies have several limitations. First, we used only one energy enhancer (CoQ₁₀). It is unknown whether other energy enhancers, such as vitamin K₂,⁶¹ can also rescue the anaesthesia-induced energy deficits and synaptic dysfunction and reduce the anaesthesia-induced cognitive deficiency. Second, we assessed the concentrations of ATP and MMP in the cortex, but not the hippocampus, of young mice because of an insufficient amount of hippocampus for measurement of ATP and for the harvest of mitochondria (for the MMP measurement). In future studies, we need to develop better methods to detect the effects of the sevoflurane anaesthesia on ATP and MMP in hippocampus of young mice.

In conclusion, in this proof-of-concept study, we demonstrated that CoQ₁₀, an energy enhancer, reduced sevoflurane-induced cognitive deficiency and rescued sevoflurane-induced reduction in the synaptic markers MMP and ATP in young mouse hippocampus. These results imply that anaesthesia causes brain energy deficits by impairing mitochondrial function, which induces synaptic dysfunction, finally leading to cognitive deficiency in young mice. These discoveries provide the basis for more research to examine the underlying mechanisms and targeted interventions for potential cognitive dysfunction after anaesthesia and surgery in children.

Authors' contributions

Conceived and designed the experiments: X.G., D.Y., Z.Y., S.G.S., L.X., X.Z.

Performed the experiments: X.G., L.H., D.Y.

Analysed the data: X.G., L.H.

Wrote the paper: X.G., L.H, S.D., X.Z.

Handling editor: Hugh C. Hemmings Jr