Secondary Logo

Journal Logo

Original Research Articles: Original Laboratory Research Report

Exogenous GM1 Ganglioside Attenuates Ketamine-Induced Neurocognitive Impairment in the Developing Rat Brain

Meng, Chen PhD*,†; Yao, Xue-qin MD*,†; Chang, Rui-jie MD*,†; Wang, Si-lu MD*,†; Wang, Xue MD*,†; Ma, Da-qing PhD; Li, Qing PhD*,†; Wang, Xian-yu PhD*,†

Author Information
doi: 10.1213/ANE.0000000000004570

Abstract

KEY POINTS

  • Question: Does monosialotetrahexosylganglioside (GM1), with the ability to attenuate neuron damage and enhance the synthesis of brain-derived neurotrophic factor (BDNF), improve the ketamine-induced neonatal anesthetic neurotoxicity?
  • Findings: The exogenous GM1 significantly ameliorated cognitive impairment, reduced the increased neuroapoptosis, and enhanced the hippocampal BNDF expression in ketamine-exposed brain, whereas the BDNF-neutralizing antibodies counteracted the exogenous GM1-induced improvement of ketamine neurotoxicity.
  • Meaning: Our data offer positive potential for the clinical use of GM1 to prevent the cognitive deficits induced by ketamine in the young per se.

Ketamine, an N-methyl-d-aspartate (NMDA) receptor antagonist, is ubiquitously used for providing anesthesia and analgesia to pediatric patients.1 Since 1999, ketamine has been reported to trigger neurodegeneration and long-term neurocognitive deficits in several preclinical investigations.2–4 In addition, ketamine, also known as “special K” with strong hallucinogenic effects, becomes increasingly involved in illegal abuse, especially among adolescents,5 and prolonged exposure to this compound is reported to cause significant damage to brain microstructure in adolescent monkeys.6 Unfortunately, the strategies to prevent/treat ketamine-induced developmental neurotoxicity is still not available.

In the development of the central nervous system, gangliosides are abundant in the brain and play key roles in numerous neural physiological processes, such as the enhancement of neural differentiation and plasticity.7 Monosialotetrahexosylganglioside (GM1), one of the most studied gangliosides, has been clinically used in humans and shown several neuroprotective properties.8,9 The exogenously administered GM1 can reduce the acute injuries of neural cells and promote the recovery processes of nerve fibers.10,11 Our previous in vitro experiment has demonstrated that GM1 directly alleviates the neurotoxicity induced by certain anesthetics in neural stem cells.12 Moreover, GM1 exhibits neurotrophic activity, which derives from its capability to enhance the synthesis and release of several neurotrophins including brain-derived neurotrophic factor (BDNF),13,14 a well-known member of the neurotrophin family of growth factors and enriched throughout the central nervous system. Dysregulation of BDNF function is reported to be deeply implicated in several neurobehavioral disorders such as depression and drug abuse.15,16 Furthermore, downregulation of BDNF and its signaling, including the downstream signaling of protein kinase B (AKT) and ERK1/2, is closely involved in ketamine-caused learning and memory impairment in the developing brain.17–19 However, to date, little is known about the effect of GM1 on the BDNF level in anesthetic-exposed immature brain and the associated developmental neurotoxicity.

Given the neuroprotective and neurotrophic activity of GM1, we hypothesize that GM1 may promote neuronal BDNF expression, thereby improving ketamine-induced neuroapoptosis and subsequent neurological deficits in the developing brain. To test this hypothesis, we first assessed the effect of GM1 on spatial learning and memory impairment in ketamine-treated young rats. This model has been widely used to study the developmental neurotoxicity of ketamine.2,4 Then to gain insights into potential therapeutic role of GM1 in ketamine-induced developmental neurotoxicity, we investigated whether BDNF signaling mediated this neuroprotective effect.

METHODS

Animals

All experimental protocols were approved by the Animal care and Use Committee of Hubei University of Medicine (Shiyan, Hubei, China), and in accordance with the relevant sections of the Animal Research: Reporting In Vivo Experiment (ARRIVE) guidelines. Sprague-Dawley rat pups at postnatal day 2 (P2) were obtained from the Institute of Laboratory Animal Science, Hubei University of Medicine. The pups were kept in a 12-hour light/dark cycle under specific pathogen-free conditions at a room temperature of 23°C ± 1°C in the Experimental Animal Center of Hubei Medical of University.

Experimental Procedures

The rat pups were tagged and randomly assigned to each group at P3 in accordance with the random number table. All pups were treated in a temperature-controlled acrylic container which maintained core body temperatures at 37°C ± 0.5°C.4 Ketamine hydrochloride (Jiangsu Hengrui Medicine, Jiangsu, China) was dissolved in saline. At P7, ketamine or control treatment was performed by 5 intraperitoneal injections (10 mL/kg each) of either ketamine (20 mg/kg) or normal saline (N.S.) at 1.5-hour intervals over a 6-hour period. To avoid hypoxia, the pups were provided with 1.5 L of oxygen per minute throughout ketamine exposure. A total of 3 in vivo experiments were conducted with 198 rats, and the number of rats in each group is shown in Figure 1.

Figure 1.
Figure 1.:
Experimental protocols and animal groups. A, Experiment 1: arterial blood gas analysis and Spo 2 monitoring to test the impact of ketamine on respiratory function in P7 rats. B, Experiment 2: behavioral analysis, apoptosis detection (TUNEL-based assay), and protein expression studies (western blotting) to investigate the effects of GM1 on ketamine-caused cognitive impairment in newborn rats. C, Experiment 3: the same detection used in experiment 2 to test the role of BDNF in mediating the effects of GM1 on cognitive impairment caused by ketamine. Numbers of animals in each experiment are also shown. D–F, Spo 2 from the neck (D), Pao 2 (E), and Paco 2 (F) in P7 rats. Data are shown as mean ± standard deviation from 3 independent samples. Two-way ANOVA showed no significant interaction of ketamine and time on Spo 2 (ketamine: F 1,16 = 0.571, P = .461; time: F 3,16 = 0.191, P = .901; interaction: F 3,16 = 1.714, P = .204), Pao 2 (ketamine: F 1,16 = 0.015, P = .924; time: F 3,16 = 0.407, P = .750; interaction: F 3,16 = 0.183, P = .902), and Paco 2 (ketamine: F 1,16 = 1.251, P = .337; time: F 3,16 = 2.824, P = .083; interaction: F 3,16 = 0.924, P = .463). ANOVA indicates analysis of variance; BDNF, brain-derived neurotrophic factor; GM1, monosialotetrahexosylganglioside; N.S., normal saline; P7, postnatal day 7; Pao 2, partial pressure of oxygen; Paco 2, partial pressure of carbon dioxide; Spo 2, pulse oxygen saturation; TUNEL, terminal deoxynucleotidyl transferase–mediated 2′-deoxyuridine 5′-triphosphate nick end labeling; WB, Western blot.

In experiment 1 (Figure 1A), the P7 rat pups were divided into 2 groups: (1) Control group and (2) ketamine group. The pups were subjected to monitoring of the pulse oxygen saturation (Spo2) using a pulse oximeter (Philips, Eindhoven, the Netherlands) placed on the neck20 at 5 minutes, 2, 4, and 6 hours after the first administration of ketamine or saline. On completion of each Spo2 monitoring, carotid artery blood samples were collected immediately for blood gas analysis.

Experiment 2 (Figure 1B) was designed to determine the effect of the exogenous GM1 on ketamine-induced developmental neurotoxicity. The rat pups were randomly divided into 4 groups: (1) control + N.S. group and (2) control + GM1 group: pups receiving normal saline or GM1 were treated with normal saline (control conditions) for 6 hours; (3) ketamine + N.S. group and (4) ketamine + GM1 group: pups receiving normal saline or GM1 were treated with ketamine for 6 hours. Exogenous GM121 (30 mg/kg; Santa Cruz Biotechnology, Santa Cruz, CA) or normal saline was administered intraperitoneally at P5, P6, and P7, and the last GM1 administration was 6 hours after ketamine exposure. For behavioral tests, control + N.S. group included 8 male and 4 female rats, ketamine + N.S. group included 5 male and 7 female rats, control + GM1 group included 6 male and 6 female rats, and ketamine + GM1 group included 7 male and 5 female rats.

In experiment 3 (Figure 1C), BDNF antibodies were used to test the role of BDNF in mediating the effects of GM1 neuroprotection, and rat pups were divided into 4 groups: (1) control group, (2) ketamine + GM1 group, (3) ketamine + GM1 + immunoglobulin Y (IgY) group, and (4) ketamine + GM1 + anti-BDNF group. To neutralize BDNF in the brain, BDNF antibodies (2 μg per rat; anti-BDNF; Promega, Madison, WI) were administered 30 minutes before GM1 treatment (Figure 1C). A sterile 26-G Hamilton microsyringe (Hamilton Company, Reno, NV) was used to intranasally administer 2-μL drops of BDNF antibodies diluted in artificial cerebrospinal fluid (0.2 μg/μL) to alternating nostrils at 2-minute interval as performed previously by our group.22 Drops were placed at the opening of the nostril, allowing the pups to snort each drop into the nasal cavity. IgY (R&D Systems, Minneapolis, MN) was used as the isotype control. A total of 10 μL of dose solution was administered over a course of 5 minutes. For behavioral tests, control group included 7 male and 5 female rats, ketamine + GM1 group included 6 male and 6 female rats, ketamine + GM1 + IgY group included 7 male and 5 female rats, and ketamine + GM1 + anti-BDNF group included 5 male and 7 female rats.

TUNEL Assay

The hippocampal cell apoptosis in P8 rat pups was detected using terminal deoxynucleotidyl transferase–mediated 2′-deoxyuridine 5′-triphosphate nick end labeling (TUNEL) assay (Supplemental Digital Content, Data, http://links.lww.com/AA/C975). The TUNEL assay was applied in experiments 2 and 3.

Morris Water Maze

Morris water maze (MWM) test was performed to assess the spatial learning and memory according to previously described methods17,23 (Supplemental Digital Content, Data, http://links.lww.com/AA/C975). For each rat, the escape latency was measured at P32–P36 and the platform-crossing times were recorded at P37. The MWM test was used in experiments 2 and 3.

Western Blot

The total protein from hippocampal tissues in P8 rats was extracted by using a commercially available kit (KGP250; Nanjing Keygen Biotech Co Ltd, Nanjing, China). The protein concentrations were measured by the bicinchoninic acid protein assay (Thermo Scientific, Fremont, CA). Equal amount (25 μg) of protein extracts was separated using sodium dodecyl sulfate-polyacrylaminde gel electrophoresis (SDS-PAGE) and then transferred onto polyvinylidene difluoride membranes (Millipore, Milford, MA) by electrophoresis. The membranes were blocked with 5% nonfat milk in TBST (0.1% Tween 20 in Tris-buffered saline) for 1 hour at room temperature, and were incubated overnight at 4°C with rabbit polyclonal anti-β-actin (1:2000; Santa Cruz Biotechnology), rabbit monoclonal anticaspase 3 (1:1000; Cell Signaling Technology, Beverly, MA), rabbit monoclonal anticleaved-caspase 3 (1:1000; Cell Signaling Technology), rabbit monoclonal anti-AKT (1:1000; Cell Signaling Technology), rabbit monoclonal anti-p-AKT (1:2000; Cell Signaling Technology), rabbit monoclonal anti-ERK (1:1000; Cell Signaling Technology), rabbit monoclonal anti-p-ERK (1:1000; Cell Signaling Technology) or rabbit monoclonal anti-BDNF (1:1000; Abcam, Shanghai, China) followed by horseradish peroxidase–conjugated goat antirabbit IgG antibody (1:4000, Proteintech Group, Chicago, IL). The protein bands were visualized and photographed by a Bio-Rad Imaging System (Bio-Rad, Pleasanton, CA), and quantified by Image Lab Image acquisition and analysis software (Bio-Rad). Western blot analysis was used in experiments 2 and 3.

Statistical Analysis

All statistical analyses were performed by GraphPad Prism software package (version 5, Graphpad Software Inc, CA). The d’Agostino and Pearson omnibus normality test was used to assess the Gaussian distribution of all data. Data were expressed as mean ± standard deviation (SD). In experiment 1, 2-way analysis of variance (ANOVA; factors: ketamine and time) was applied to evaluate the difference of arterial blood gas analysis. In experiment 2, 2-way repeated measure ANOVA (factors: group and time) followed by Bonferroni posthoc test was used to evaluate the differences in the escape latency in the MWM test. The factor of group includes the groups of control + N.S. and ketamine + N.S. in Figure 2B or the groups of control + GM1 and ketamine + GM1 in Figure 2C. A 2-way ANOVA (factors: ketamine and GM1) followed by Bonferroni posthoc test was used to evaluate the differences in the platform-crossing times, TUNEL staining analysis, and relative densities of western blots. In experiment 3, a 2-way repeated measure ANOVA (factors: antibody and time) was applied to evaluate the differences in the escape latency in the MWM test. One-way ANOVA followed by Bonferroni posthoc test was used for other multiple comparisons. The P values reported throughout were all adjusted with a Bonferroni correction. A 2-tailed P < .05 was considered to be statistically significant.

Figure 2.
Figure 2.:
GM1 improved the ketamine-induced cognitive impairment in young rats. A, Representative traces of the paths swum by rats starting in the right upper quadrant in MWM on P36. The red cycle represents the platform. The red and blue points represent the starting and ending position of the rat’s behavior, respectively. B, Escape latency of the young rats in control + N.S. or ketamine + N.S. group in MWM at P32–P36. Data represent the mean ± standard deviation for 12 rats in each group. Two-way repeated measure ANOVA showed that there was a significant interaction between group (control + N.S. and ketamine + N.S.) and time (P32–P36) on escape latency of MWM test (group: F 1,11 = 28.890, P < .001; time: F 4,44 = 30.130, P < .001; interaction: F 4,44 = 4.718, P = .003). Bonferroni posthoc test revealed that the rats in ketamine + N.S. group had a significantly prolonged escape latency compared with those in control + N.S. at P34–P36. **P = .006 for P34; **P = .002 for P35; and **P = .005 for P36. C, Two-way repeated measure ANOVA showed that no significant interaction of group (control + GM1 and ketamine + GM1) and time on escape latency of MWM in the rats pretreated with GM1 (group: F 1,11 = 2.327, P = .16; time: F 4,44 = 42.580, P < .001; interaction: F 4,44 = 1.727, P = .161). D, Platform-crossing times of young rats in MWM at P37. GM1 markedly increased the platform-crossing times of ketamine-exposed rats (ketamine + GM1 versus ketamine + N.S.). Two-way ANOVA with Bonferroni posthoc test showed that there was a significant interaction of ketamine versus GM1 on platform-crossing times of MWM test (ketamine: F 1,44 = 22.970, P < .001; GM1: F 1,44 = 1.115, P = .297; interaction: F 1,44 = 9.034, P = .004). Data represent the mean ± standard deviation for 12 rats in each group. ***P < .001 versus control + N.S.; #P < .05 versus ketamine + N.S. All P values were adjusted with a Bonferroni correction. ANOVA indicates analysis of variance; GM1, monosialotetrahexosylganglioside; MWM, Morris water maze; N.S., normal saline; P36, postnatal day 36.

The calculation of sample size in our studies was referred to a previous research,24 and was performed with G*Power 3.1.9.2 (Http://www.gpower.hhu.de/en.html). The sample number was 12 in each behavioral study group, 6 in every group for western blot and TUNEL staining, and 3 in each group for analysis of arterial blood gas and Spo2. By our preliminary test (n = 5), the escape latency of P36 rats in control and ketamine groups was 26.11 ± 18.67 and 55.76 ± 23.41 seconds, respectively. According to Lu et al’s study,24 using 2-tailed Student t test with α level at .05 and power (1 - β) at .9, we calculated the minimal sample size for MWM test to be 12. For TUNEL staining, our preliminary test showed that the hippocampal TUNEL-positive cells per field in the control and ketamine groups were 3.55 ± 2.06 and 9.66 ± 3.23, respectively. For western blot analysis, the relative densities of cleaved-caspase 3/β-actin in the control and ketamine groups in the preliminary test were 0.25 ± 0.09 and 0.58 ± 0.18. A sample size of 6 rat pups in each group for western blot and TUNEL staining will provide a statistical power of >0.9. The studies of arterial blood gas analysis were quantified on the basis of our previous experience and repeated 3 times.

RESULTS

Effect of Ketamine on Respiratory Function in P7 Rats

To exclude the interference of ketamine-induced hypoxia, we estimated the oxygenated parameters in P7 rats. Compared to normal saline (control), ketamine did not affect the levels of Spo2 (ketamine: F1,16 = 0.571, P = .461; 2-way ANOVA), Pao2 (ketamine: F1,16 = 0.015, P = .924; 2-way ANOVA), and Paco2 (ketamine: F1,16 = 1.251, P = .337; 2-way ANOVA) in P7 rats at different time points (Figure 1D–F). None of the ketamine-by-time interactions were significant (all P > .05).

Effect of Exogenous GM1 on Ketamine-Induced Cognitive Impairment in Young Rats

To investigate the effect of GM1 on ketamine-induced cognitive impairment, rat pups were given exogenous GM1 or N.S. twice before and once after a 6-time ketamine challenge (Figure 1B). Then the cognitive functions were tested in the MWM. Figure 2A depicts the representative traces of the paths swum by P36 rats. As shown in Figure 2B, when compared with the control + N.S. rats, the ketamine + N.S. rats had significantly longer escape latency from P34 to P36 (P34, P = .006; P35, P = .002; P36, P = .005), suggesting that in the rats pretreated with N.S., ketamine induced cognitive impairment tested in MWM. However, in the rats pretreated with exogenous GM1 (Figure 2C), ketamine did not affect the escape latency (group: F1,11 = 2.327, P = .16, 2-way repeated measure ANOVA). Our results (Figure 2D) also showed that when compared with the rats in control + N.S. group, the ketamine + N.S. rats presented lower platform-crossing times (7.00 ± 1.71 vs 3.00 ± 2.22 times, mean ± SD; P < .001). However, no significant decrease in the platform-crossing times was noted between control + GM1 and ketamine + GM1 group (6.00 ± 1.48 vs 5.40 ± 1.53 times, mean ± SD; P > .99), and ketamine + GM1 group showed more platform-crossing times than ketamine + N.S. group (5.40 ± 1.53 vs 3.00 ± 2.22 times, mean ± SD; P = .041).

Effect of Exogenous GM1 on Ketamine-Induced Hippocampal Apoptosis in Young Rats

Figure 3.
Figure 3.:
GM1 ameliorated ketamine-induced apoptosis in hippocampus. A, Representative images of hippocampal region stained for TUNEL-positive nuclei (green) in the P8 brain. Scale bar = 50 μm. B, Quantitation of TUNEL-positive cells. Two-way ANOVA with Bonferroni multiple comparisons showed that there was a significant interaction of ketamine and GM1 on TUNEL staining (ketamine: F 1,20 = 28.240, P < .001; GM1: F 1,20 = 4.902, P = .039; interaction: F 1,20 = 4.902, P = .039). Data are shown as mean ± standard deviation for 6 rats in each group. ***P < .001 for ketamine + N.S. versus control + N.S.; #P = .032 for ketamine + GM1 versus ketamine + N.S. C, Total and cleaved-caspase 3 in P8 hippocampus were tested by western blot. D, Densitometric analysis of cleaved-caspase 3/β-actin. Two-way ANOVA with Bonferroni multiple comparisons showed a significant interaction of ketamine and GM1 on cleaved-caspase 3 expression (ketamine: F 1,20 = 90.240, P < .001; GM1: F 1,20 = 5.152, P = .034; interaction: F 1,20 = 4.417, P = .048). Data are shown as mean ± standard deviation for 6 rats in each group. ***P < .001 for ketamine + N.S. versus control + N.S.; ***P < .001 for ketamine + GM1 versus control + GM1; #P = .035 for ketamine + GM1 versus ketamine + N.S. All P values were adjusted with a Bonferroni correction. ANOVA indicates analysis of variance; DAPI, 4′,6-diamidino-2-phenylindole; GM1, monosialotetrahexosylganglioside; N.S., normal saline; P8, postnatal day 8; TUNEL, terminal deoxynucleotidyl transferase–mediated 2′-deoxyuridine 5′-triphosphate nick end labeling.

To understand the role of the exogenous GM1 in ketamine-induced neurodegeneration, we determined the cellular apoptosis and cleaved-caspase 3 expression in the hippocampus. Figure 3A represents TUNEL staining results of the hippocampus from P8 rats. The ketamine-exposed rats experienced a significant increase in hippocampal TUNEL-positive cells (Figure 3B) and cleaved-caspase 3 expression (Figure 3D) when compared with the controls (ketamine + N.S. versus control + N.S.; both P < .001). However, the ketamine + GM1 rat pups showed remarkably less TUNEL staining and cleaved-caspase 3 expression in hippocampus than the ketamine + N.S. pups (P = .032 and P = .035, respectively).

Effect of Exogenous GM1 on BDNF Signaling in Ketamine-Exposed Rat Pups

Figure 4.
Figure 4.:
GM1 promoted BDNF expression and activated AKT and ERK1/2 phosphorylation. Representative images of western blot (A) and densitometric analysis of BDNF/β-actin (B). There was a significant interaction of ketamine and GM1 on hippocampal expression of BDNF (ketamine: F 1,20 = 41.020, P < .001; GM1: F 1,20 = 2.806, P = .109; interaction: F 1,20 = 8.165, P = .010; 2-way ANOVA). ***P < .001 for ketamine + N.S. versus control + N.S.; #P = .027 for ketamine + N.S. versus ketamine + GM1. C, Densitometric analysis of phosphorylated/total ERK1/2. There was a significant interaction of ketamine and GM1 on p-ERK1/2 expression (ketamine: F 1,20 = 52.210, P < .001; GM1: F 1,20 = 6.162, P = .022; interaction: F 1,20 = 3.102, P = .094; 2-way ANOVA). ***P < .001 for ketamine + N.S. versus control + N.S., **P = .006 for ketamine + GM1 versus ketamine + N.S.; #P = .042 for ketamine + N.S. versus ketamine + GM1. D, Densitometric analysis of phosphorylated/total AKT. A significant interaction of ketamine and GM1 on p-AKT expression was noted (ketamine: F 1,20 = 62.330, P < .001; GM1: F 1,20 = 2.463, P = .132; interaction: F 1,20 = 7.220, P = .014; 2-way ANOVA). ***P < .001 for ketamine + N.S. versus control + N.S., **P = .009 for ketamine + GM1 versus ketamine + N.S.; #P = .039 for ketamine + N.S. versus ketamine + GM1. Values are mean ± standard deviation for 6 rats in each group. Two-way ANOVA with Bonferroni multiple comparisons. All P values were adjusted with a Bonferroni correction. AKT indicates protein kinase B; ANOVA, analysis of variance; BDNF, brain-derived neurotrophic factor; GM1, monosialotetrahexosylganglioside; N.S., normal saline; P8, postnatal day 8.

Figure 4A represents the western blotting images of BDNF signaling proteins in the hippocampus from P8 rats. The densitometric analysis showed that the levels of BDNF notably decreased in the ketamine-challenged rat pups compared with the controls (ketamine + N.S. versus control + N.S., P < .001; Figure 4B), but GM1 treatment significantly inhibited the ketamine-induced BDNF decrease (ketamine + GM1 versus ketamine + N.S.; P = .027). In addition, the results also showed that when compared with ketamine + N.S. group, ketamine + GM1 group exhibited higher levels of ERK1/2 (P = .042; Figure 4C) and AKT (P = .039; Figure 4D) phosphorylation in the hippocampus.

Effect of BDNF Antibody on GM1-Treated Rat Pups Exposed to Ketamine

Figure 5.
Figure 5.:
Inhibition of BDNF almost abolished the effect of GM1 on ketamine-exposed young rats. A and B, Western blot and densitometric analysis of BDNF in hippocampus of P8 rats treated with ketamine plus GM1 in the presence of different doses of BDNF antibodies. One-way ANOVA with Bonferroni multiple comparisons showed that there was a significant effect of BDNF antibodies on hippocampal BDNF expression (F 3,20 = 27.566; P < .001). Values are mean ± standard deviation for 6 rats in each group. ***P < .001 versus ketamine + GM1. C, Representative traces of the paths swum by P36 rats starting in the right upper quadrant of MWM. D, Escape latency of young rats in MWM at P32–P36. Two-way repeated-measures ANOVA with Bonferroni multiple comparisons showed that there was a significant interaction of antibody versus time on escape latency (antibody: F 3,44 = 12.830, P < .001; time: F 4,176 = 90.422, P < .001; interaction: F 12,176 = 2.231, P = .018). Values are mean ± standard deviation for 12 rats in each group. *P = .027 versus ketamine + GM1 + IgY at P36. E, Platform-crossing times of young rats in MWM at P37. One-way ANOVA posthoc Bonferroni test showed that BDNF antibodies markedly decreased the platform-crossing times of P37 rats treated with ketamine plus GM1 (F 3,44 = 7.049; P < .001). Data are shown as mean ± standard deviation of 12 independent samples. *P = .042 for ketamine + GM1 versus control, *P = .037 for ketamine + GM1 + IgY versus control, ***P < .001 for ketamine + GM1 + anti-BDNF versus control; #P = .045 for ketamine + GM1 + anti-BDNF versus ketamine + GM1 + IgY. All P values were adjusted with a Bonferroni correction. ANOVA indicates analysis of variance; BDNF, brain-derived neurotrophic factor; GM1, monosialotetrahexosylganglioside; IgY, immunoglobulin Y; MWM, Morris water maze; P37, postnatal day 37.

Then we investigated the role of BDNF in the neuroprotective effect of GM1. Initially, we detected whether the BDNF antibodies intranasally administered could neutralize the hippocampal BDNF protein. Western blot analysis (Figure 5A, B) showed that the BDNF antibodies dose-dependently reduced the levels of hippocampal BDNF (F3,20 = 27.566, P < .001, 1-way ANOVA), and the most potent effect was observed at a dose of 100 μg/kg in our experiment. Thus the 100 μg/kg BDNF antibody was used intranasally for further study. IgY was applied as an isotype control. Figure 5C depicts the representative traces of the paths swum by P36 rats, and Figure 5D showed that when exposed to ketamine, the GM1-treated rat pups receiving BDNF antibodies experienced more severe cognitive impairment compared with their IgY-receiving counterparts (ketamine + GM1 + anti-BDNF versus ketamine + GM1 + IgY), as evidenced by longer escape latency at P36 (41.32 ± 12.37 seconds vs 25.14 ± 8.97 seconds, mean ± SD, P = .027; Figure 5D). In addition, ketamine + GM1 + anti-BDNF group had less platform-crossing times at P37 as compared with ketamine + GM1 + IgY group (2.16 ± 1.12 times vs 3.92 ± 1.97 times, mean ± SD, P = .036, 1-way ANOVA with posthoc Bonferroni test; Figure 5E).

Effect of BDNF Antibody on GM1-Induced Neuroapoptosis Improvement in Ketamine-Exposed Rat Pups

Figure 6.
Figure 6.:
Block of BDNF inhibited the effect of GM1 on ketamine-induced apoptosis in hippocampus. A and B, Representative images of hippocampal regions stained with TUNEL (A) and the quantitation of TUNEL-positive cells (B) in the P8 brain (*P = .041 for ketamine + GM1 versus control, *P = .035 for ketamine + GM1 + IgY versus control and ***P < .001 for ketamine + GM1 + anti-BDNF versus control; #P = .043 for ketamine + GM1 + anti-BDNF versus ketamine + GM1 + IgY; 1-way ANOVA). C and D, Western blot analysis of total and cleaved-caspase 3 in hippocampal tissues (C) and densitometric analysis (D) of cleaved-caspase 3/β-actin in P8 rats (**P = .006 for ketamine + GM1 versus control, **P = .009 for ketamine + GM1 + IgY versus control and ***P < .001 for ketamine + GM1 + anti-BDNF versus control; #P = .039 for ketamine + GM1 + anti-BDNF versus ketamine + GM1 + IgY; 1-way ANOVA). E–G, Western blot and densitometric analysis of ERK1/2 (*P = .021 for ketamine + GM1 versus control, *P = .011 for ketamine + GM1 + IgY versus control and ***P < .001 for ketamine + GM1 + anti-BDNF versus control; #P = .047 for ketamine + GM1 + anti-BDNF versus ketamine + GM1 + IgY; 1-way ANOVA) and AKT (*P = .023 for ketamine + GM1 versus control, *P = .011 for ketamine + GM1 + IgY versus control and ***P < .001 for ketamine + GM1 + anti-BDNF versus control; ##P = .008 for ketamine + GM1 + anti-BDNF versus ketamine + GM1 + IgY; 1-way ANOVA) in hippocampus of P8 rats. Data are shown as mean ± standard deviation of 6 independent samples. All P values were adjusted with a Bonferroni correction. AKT indicates protein kinase B; ANOVA, analysis of variance; BDNF, brain-derived neurotrophic factor; ERK1/2, extracellular signal-related kinases 1 and 2; GM1, monosialotetrahexosylganglioside; IgY, immunoglobulin Y; P8, postnatal day 8; TUNEL, terminal deoxynucleotidyl transferase–mediated 2′-deoxyuridine 5′-triphosphate nick end labeling.

We next investigated whether BDNF mediated the effect of GM1 on neuroapoptosis improvement. In this experiment, 1-way ANOVA with Bonferroni multiple comparisons showed that when compared with the IgY-treated counterparts (ketamine + GM1 + anti-BDNF versus ketamine + GM1 + IgY), the BDNF antibody-treated rat pups administered with GM1 under ketamine exposure exhibited more TUNEL-positive cells (P = .043; Figure 6A, B) and higher levels of cleaved-caspase 3 expression (P = .039; Figure 6C, D) in the hippocampus. Finally, the phosphorylation of ERK1/2 and AKT was determined after the BDNF antibody treatment. One-way ANOVA with Bonferroni multiple comparisons showed that when compared with their IgY-receiving counterparts (ketamine + GM1 + anti-BDNF versus ketamine + GM1 + IgY), the BDNF antibodies almost abolished the phosphorylation of both ERK1/2 (P = .047; Figure 6E, F) and AKT (P = .008; Figure 6E, G) in the hippocampus of exogenous GM1-treated pups exposed to ketamine.

DISCUSSION

In this therapeutic investigation of ketamine-induced neurotoxicity and cognitive impairment in the developing brain, our data demonstrated for the first time that GM1 attenuated neuroapoptosis in the hippocampus and improved neurobehavioral deficits in ketamine-exposed neonatal rats. Moreover, the neuroprotective effects of GM1 on the developing brain challenged with ketamine were mediated by the BDNF signaling pathway, as these effects were almost abolished by the selective blocking of BDNF in the brain.

It has been reported that prolonged exposure to ketamine, one of the most used drugs for pediatric anesthesia, can directly cause significant neuronal damage and neurodevelopmental disorder in the developing brain.2,4 In addition, ketamine treatment commonly results in hypoventilation and hypoxia in children,25 such condition can also lead to lifelong neurobehavioral dysfunction26 but can be easily reversed by nasal oxygen. In this study, we first detected whether ketamine administration caused hypoxia in newborn rats in an oxygen-rich atmosphere. Our data showed that the levels of Spo2, Pao2, and Paco2 were all within the normal range in ketamine-challenged rats and were highly similar with those in the control. Thus, we can completely exclude the interference of ketamine-induced hypoxia.

Various clinical studies8,27 showed that GM1, with the property of directly penetrating the blood-brain barrier reaching neurons,9,28 has been widely used in humans by intravenous administration. Apart from this, the most important reason of present study for assuming the potential therapeutic role of GM1 in anesthetic-induced neurotoxicity originated from several reports demonstrating that GM1 features important physiological properties including limiting neuronal damage and promotion of neurotrophin release.10,14 GM1 has been proven to possess multiple pharmacological effects on neurological disorders, such as improvement in cerebral ischemia-reperfusion, Alzheimer disease, and depression.21,29,30 In this study, our data showed that GM1 treatment reduced the escape latency, increased platform crossings in the MWM test, and improved the learning and memory ability of ketamine-challenged young rats. To our knowledge, this study presented the first evidence showing that GM1 exerts beneficial effects against anesthetic-induced cognitive deficits in the developing brain.

Previous studies reported that neuroapoptosis is a key mechanism underlying ketamine-induced neurodevelopmental damage,31 and inhibition of neuroapoptosis can improve neurobehavioral dysfunction in the ketamine-induced neonatal neurodamage model.4 GM1, with an activity similar to that of neurotrophins, can reduce acute nerve cell damage and enhance the survival of different types of neuron.9 In view of these discoveries, we focused on the effect of GM1 on the ketamine-induced apoptosis to investigate its potential mechanisms. Our results showed that GM1 significantly reduced cleaved-caspase 3 expression in the ketamine-exposed hippocampus, accompanied by the reduction of TUNEL-positive cells in the same area. Previous studies have identified that ketamine-induced apoptotic activation occurs primarily in hippocampal neurons which are highly related to learning and memory tasks.2 On this foundation, our findings suggested that GM1 attenuated the ketamine-induced neuroapoptosis in the developing brain hippocampus, resulting in the neurobehavioral improvement.

Studies have reported that special block of NMDA receptor can result in sustained depletion of BDNF expression in the developing brain.32 As a classical NMDA receptor antagonist, ketamine is demonstrated to directly lower BDNF levels in immature brain and neural cells.18,33 Even the exposure of ketamine to pregnant rats can cause significant BDNF downregulation in the hippocampus of their offspring pups.34,35 Consistent with these studies, we observed that a 6-hour treatment of ketamine at the dose of 20 mg/kg produced a notable decrease in the BDNF expression in neonatal hippocampus. However, a contradiction also exists between ketamine stimulation and BDNF secretion. Several studies reported that in depression treatment, ketamine increased BDNF expression in the hippocampus,36,37 showing an opposite effect. This contradiction may be mainly due to the different administration dose of ketamine. Ketamine can dose-dependently induce neuronal cell death during development, and the 20 mg/kg ketamine dose used in our experiment was proven to lead to significant neuronal damage in the developing hippocampus.38 In the depression treatment, a subanesthetic ketamine dose of 10 mg/kg or less, which is lower than what we used, was applied.36,37 This lower dose of ketamine hardly caused observable immature brain damage.38 Besides, the age of rats receiving ketamine is markedly different. Unlike the newborn rats used in our experiments, adult rats are usually studied for depression investigation.36,37 The different ketamine doses and rat age may be primarily responsible for the distinct results, leaving us with additional questions regarding whether different doses of ketamine administrated to brain at different stages can lead to different BDNF expression patterns and different behavioral outcomes. This warrants further investigation.

The exogenous administration of GM1 has been shown to promote the release of BDNF from hippocampal neurons in several neuropathological disorder models such as in depressed rats.21 Similarly, our data showed that GM1 treatment significantly increased the hippocampal BDNF expression in ketamine-exposed newborn rats. BDNF is well known for its critical function in brain structural remodeling, neuronal survival, and the capacity for learning and memory.15,39 In our study, blocking of BDNF by antibodies almost abolished the exogenous GM1-caused improvement of hippocampal apoptosis and neurobehavioral deficits in the ketamine-challenged rat pups. Our results are similar with those of a previous study showing that BDNF can mediate the neuroimprovement function of GM1,21 suggesting that BDNF potentially plays a key role in mediating the protective effect of exogenous GM1 against ketamine neurotoxicity. However, the mediation analysis was not performed because such analyses are complex and beyond the scope of this study. So we cannot quantify how much of the protective effect of GM1 due to BDNF mediation. This is a limitation of our experiment.

In this study, we also observed that in parallel with the lowered BDNF expression, the phosphorylation of AKT and ERK1/2 was significantly inhibited in the ketamine-challenged neonatal hippocampus, in accordance with previous reports showing that downregulation of AKT and ERK1/2 phosphorylation is implicated in developmental neurotoxicity of ketamine.4,17,19 Both AKT and ERK1/2 can be activated by BDNF signaling,39 promote cellular survival after injury, and be inactivated in hippocampus under high doses of ketamine exposure. ERK1/2 facilitates the NMDA receptor–induced neuroprotection,39 and uncoupling of NMDA receptor from ERK1/2 signaling leads to massive apoptotic depletion of neurons in the developing brain.32 As for AKT, its upregulated activation is closely related to the improvement of ketamine-induced neuroapoptotic activity and is able to directly recover ketamine-induced spatial learning and memory deficits.4,40 In our experiment, the ketamine-induced downregulated phosphorylation of AKT and ERK1/2 was markedly reversed by the exogenous GM1 treatment, and this effect could be inhibited by the BDNF antibody. Again, because of the lack of statistical mediation analysis, the function of BDNF in the activation of AKT and ERK1/2 was not identified quantitatively. Collectively, these findings suggested that BDNF signaling, including AKT and ERK1/2, may play a contributing role in the protective effects of GM1 on ketamine-induced hippocampal apoptosis and neurobehavioral deficits in developing brain, therefore causing GM1 to exhibit neurotrophin-like activity.

This study features several limitations. We opted not to test the activation level of TrkB after the BDNF treatment. BDNF directly activates TrkB, and subsequently, the downstream signaling cascades including ERK1/2 and AKT to exert its neuroprotective functions.39 Given their close association with ketamine-caused cognitive deficits in the brain, we directly focused on ERK1/2 and AKT. BDNF-ERK1/2 signaling is also deeply involved in long-term potentiation, which is important in certain forms of learning and memory.39 Given the technical constraints, we opted not to examine whether long-term potentiation is involved in the neuroprotective effects of exogenous GM1 in developing brain. This issue warrants further investigation. In addition, there is a growing realization that sex is an important biological variable in rodent research in anesthetic-induced developmental neurotoxicity. But to our knowledge, there has been no systematic study on identifying the effect of sex factor on developmental cognitive dysfunction caused by the exposure to ketamine at the anesthetic dosage. This issue was not addressed in the present study, either, and will be our next research goal.

In summary, this study showed that GM1 exerts a protective effect on the developing brains by inducing a high-level expression of BDNF and enhancing the activation of its downstream signaling of AKT and ERK1/2 in the hippocampus, potentially mitigating hippocampal apoptosis and improving the cognitive deficits in learning and memory, and resultantly protecting the developing brain against ketamine-induced neurotoxicity. Therefore, our study may offer the positive therapeutic potential of GM1 in treating/preventing the cognitive impairments induced by ketamine in the young.

ACKNOWLEDGMENTS

We express special gratitude to the School of Basic Medical Sciences at Hubei University of Medicine for providing the experimental platform.

DISCLOSURES

Name: Chen Meng, PhD.

Contribution: This author helped design the study, conduct the experiment (behavioral test), analyze the data, and write the final manuscript.

Name: Xue-qin Yao, MD.

Contribution: This author helped perform the experiments (treatment, behavioral test, and Western blot) and collect and analyze the data.

Name: Rui-jie Chang, MD.

Contribution: This author helped perform the experiments (Western blot) and analyze the data.

Name: Si-lu Wang, MD.

Contribution: This author helped perform the experiments (treatment and TUNEL assay) and analyze the data.

Name: Xue Wang, MD.

Contribution: This author helped perform the experiments (TUNEL assay) and analyze the data.

Name: Da-qing Ma, PhD.

Contribution: This author helped revise the manuscript.

Name: Qing Li, PhD.

Contribution: This author helped design the study, analyze the data, and write the manuscript.

Name: Xian-yu Wang, PhD.

Contribution: This author helped design the study, analyze the data, and modify the manuscript.

This manuscript was handled by: Gregory J. Crosby, MD.

FOOTNOTES

GLOSSARY

AKT = = protein kinase B

ANOVA = = analysis of variance

ARRIVE = = Animal Research: Reporting In Vivo Experiment

BDNF = = brain-derived neurotrophic factor

ERK = = extracellular signal-regulated protein kinase

DAPI = = 4′,6-diamidino-2-phenylindole

GM1 = = monosialotetrahexosylganglioside

IgY = = immunoglobulin Y

MWM = = Morris water maze

NMDA = = N-methyl-d-aspartate

N.S. = = normal saline

P2 = = postnatal day 2

Paco2 = = partial pressure of carbon dioxide

Pao2 = = partial pressure of oxygen

SD = = standard deviation

SDS-PAGE = = sodium dodecyl sulfate-polyacrylaminde gel electrophoresis

Spo2 = = pulse oxygen saturation

TBST = = tris-buffered saline

TUNEL = = terminal deoxynucleotidyl transferase–mediated 2′-deoxyuridine 5′-triphosphate nick end labeling

WB = = western blot

REFERENCES

1. Green SM, Coté CJ. Ketamine and neurotoxicity: clinical perspectives and implications for emergency medicine. Ann Emerg Med. 2009;54:181–190.
2. Soriano SG, Liu Q, Li J, et al. Ketamine activates cell cycle signaling and apoptosis in the neonatal rat brain. Anesthesiology. 2010;112:1155–1163.
3. Ikonomidou C, Bosch F, Miksa M, et al. Blockade of NMDA receptors and apoptotic neurodegeneration in the developing brain. Science. 1999;283:70–74.
4. Liu JR, Liu Q, Li J, et al. Noxious stimulation attenuates ketamine-induced neuroapoptosis in the developing rat brain. Anesthesiology. 2012;117:64–71.
5. McCambridge J, Winstock A, Hunt N, Mitcheson L. 5-Year trends in use of hallucinogens and other adjunct drugs among UK dance drug users. Eur Addict Res. 2007;13:57–64.
6. Li Q, Shi L, Lu G, et al. Chronic ketamine exposure causes white matter microstructural abnormalities in adolescent cynomolgus monkeys. Front Neurosci. 2017;11:285.
7. Sandhoff K, Harzer K. Gangliosides and gangliosidoses: principles of molecular and metabolic pathogenesis. J Neurosci. 2013;33:10195–10208.
8. Geisler FH, Coleman WP, Grieco G, Poonian D; Sygen Study Group. The sygen multicenter acute spinal cord injury study. Spine (Phila Pa 1976). 2001;26:S87–S98.
9. Aureli M, Mauri L, Ciampa MG, et al. GM1 ganglioside: past studies and future potential. Mol Neurobiol. 2016;53:1824–1842.
10. Lipartiti M, Lazzaro A, Zanoni R, et al. Monosialoganglioside GM1 reduces NMDA neurotoxicity in neonatal rat brain. Exp Neurol. 1991;113:301–305.
11. Facci L, Leon A, Toffano G, Sonnino S, Ghidoni R, Tettamanti G. Promotion of neuritogenesis in mouse neuroblastoma cells by exogenous gangliosides. Relationship between the effect and the cell association of ganglioside GM1. J Neurochem. 1984;42:299–305.
12. Lu J, Yao XQ, Luo X, et al. Monosialoganglioside 1 may alleviate neurotoxicity induced by propofol combined with remifentanil in neural stem cells. Neural Regen Res. 2017;12:945–952.
13. Valdomero A, Perondi MC, Orsingher OA, Cuadra GR. Exogenous GM1 ganglioside increases accumbal BDNF levels in rats. Behav Brain Res. 2015;278:303–306.
14. Lim ST, Esfahani K, Avdoshina V, Mocchetti I. Exogenous gangliosides increase the release of brain-derived neurotrophic factor. Neuropharmacology. 2011;60:1160–1167.
15. Logrip ML, Barak S, Warnault V, Ron D. Corticostriatal BDNF and alcohol addiction. Brain Res. 2015;1628:60–67.
16. Zhao H, Alam A, San CY, et al. Molecular mechanisms of brain-derived neurotrophic factor in neuro-protection: recent developments. Brain Res. 2017;1665:1–21.
17. Huang L, Liu Y, Jin W, Ji X, Dong Z. Ketamine potentiates hippocampal neurodegeneration and persistent learning and memory impairment through the PKCγ-ERK signaling pathway in the developing brain. Brain Res. 2012;1476:164–171.
18. Goulart BK, de Lima MN, de Farias CB, et al. Ketamine impairs recognition memory consolidation and prevents learning-induced increase in hippocampal brain-derived neurotrophic factor levels. Neuroscience. 2010;167:969–973.
19. Zuo D, Lin L, Liu Y, et al. Baicalin attenuates ketamine-induced neurotoxicity in the developing rats: involvement of PI3K/Akt and CREB/BDNF/Bcl-2 Pathways. Neurotox Res. 2016;30:159–172.
20. Uchima Koecklin KH, Hiranuma M, Kato C, et al. Unilateral nasal obstruction during later growth periods affects craniofacial muscles in rats. Front Physiol. 2016;7:669.
21. Jiang B, Song L, Wang C, et al. Antidepressant-like effects of gm1 ganglioside involving the bdnf signaling cascade in mice. Int J Neuropsychoph. 2016;19:pyw046.
22. Meng C, Zhang JC, Shi RL, Zhang SH, Yuan SY. Inhibition of interleukin-6 abolishes the promoting effects of pair housing on post-stroke neurogenesis. Neuroscience. 2015;307:160–170.
23. Vorhees CV, Williams MT. Morris water maze: procedures for assessing spatial and related forms of learning and memory. Nat Protoc. 2006;1:848–858.
24. Lu H, Liufu N, Dong Y, et al. Sevoflurane acts on ubiquitination-proteasome pathway to reduce postsynaptic density 95 protein levels in young mice. Anesthesiology. 2017;127:961–975.
25. Langhan ML, Chen L, Marshall C, Santucci KA. Detection of hypoventilation by capnography and its association with hypoxia in children undergoing sedation with ketamine. Pediatr Emerg Care. 2011;27:394–397.
26. Ziemka-Nalecz M, Jaworska J, Zalewska T. Insights into the neuroinflammatory responses after neonatal hypoxia-ischemia. J Neuropathol Exp Neurol. 2017;76:644–654.
27. Bilotta F, Gelb AW, Stazi E, Titi L, Paoloni FP, Rosa G. Pharmacological perioperative brain neuroprotection: a qualitative review of randomized clinical trials. Br J Anaesth. 2013;110Suppl 1i113–i120.
28. Ghidoni R, Fiorilli A, Trinchera M, Venerando B, Chigorno V, Tettamanti G. Uptake, cell penetration and metabolic processing of exogenously administered GM1 ganglioside in rat brain. Neurochem Int. 1989;15:455–465.
29. Zhang J, Fang X, Zhou Y, et al. The possible damaged mechanism and the preventive effect of monosialotetrahexosylganglioside in a rat model of cerebral ischemia-reperfusion injury. J Stroke Cerebrovasc Dis. 2015;24:1471–1478.
30. Yang R, Wang Q, Min L, Sui R, Li J, Liu X. Monosialoanglioside improves memory deficits and relieves oxidative stress in the hippocampus of rat model of Alzheimer’s disease. Neurol Sci. 2013;34:1447–1451.
31. Vutskits L, Xie Z. Lasting impact of general anaesthesia on the brain: mechanisms and relevance. Nat Rev Neurosci. 2016;17:705–717.
32. Hansen HH, Briem T, Dzietko M, et al. Mechanisms leading to disseminated apoptosis following NMDA receptor blockade in the developing rat brain. Neurobiol Dis. 2004;16:440–453.
33. Zuo D, Sun F, Cui J, et al. Alcohol amplifies ketamine-induced apoptosis in primary cultured cortical neurons and PC12 cells through down-regulating CREB-related signaling pathways. Sci Rep. 2017;7:10523.
34. Li X, Guo C, Li Y, et al. Ketamine administered pregnant rats impair learning and memory in offspring via the CREB pathway. Oncotarget. 2017;8:32433–32449.
35. Zhao T, Li Y, Wei W, Savage S, Zhou L, Ma D. Ketamine administered to pregnant rats in the second trimester causes long-lasting behavioral disorders in offspring. Neurobiol Dis. 2014;68:145–155.
36. Ma Z, Zang T, Birnbaum SG, et al. TrkB dependent adult hippocampal progenitor differentiation mediates sustained ketamine antidepressant response. Nat Commun. 2017;8:1668.
37. Choi M, Lee SH, Park MH, Kim YS, Son H. Ketamine induces brain-derived neurotrophic factor expression via phosphorylation of histone deacetylase 5 in rats. Biochem Biophys Res Commun. 2017;489:420–425.
38. Zou X, Patterson TA, Sadovova N, et al. Potential neurotoxicity of ketamine in the developing rat brain. Toxicol Sci. 2009;108:149–158.
39. Minichiello L. TrkB signalling pathways in LTP and learning. Nat Rev Neurosci. 2009;10:850–860.
40. Zhang YH, Zhang J, Song JN, et al. The PI3K-AKT-mTOR pathway activates recovery from general anesthesia. Oncotarget. 2016;7:40939–40952.

Supplemental Digital Content

Copyright © 2019 International Anesthesia Research Society