https://orcid.org/0009-0007-9922-5603
Figures
Abstract
Anesthetic and sedative drugs are small compounds known to bind to hundreds of proteins. One intriguing binding partner of propofol is the motor domain of a neuronal mitochondrial transport kinesin, kif5A. Here, we used zebrafish wild type (WT) and kif5Aa knockout (KO) larval behavioral assays to assess anesthetic sensitivity and combined that with zebrafish primary neuronal cell culture to probe for alterations in mitochondrial motility. We found that the loss of kif5Aa increases behavioral sensitivity to propofol and etomidate, with etomidate hypersensitivity greater than propofol. In contrast, kif5Aa KO animals were resistant to the behavioral effects of dexmedetomidine. Finally, WT and kif5Aa KO larvae responded similarly to the behavioral effects of ketamine. Propofol inhibited the anterograde motility of mitochondria in WT zebrafish neurons, while etomidate inhibited mitochondrial motility in both anterograde and retrograde directions; neither drug altered mitochondrial motility in the kif5Aa knockout (KO) neurons. In contrast, dexmedetomidine enhanced retrograde mitochondrial motility in both WT and kif5Aa KO animals. Finally, ketamine had little significant effect on mitochondrial motility in either mutant or WT animals. These data demonstrate that each anesthetic/sedative drug affects the motor protein machinery uniquely and is associated with unique changes in behavior. Understanding how different anesthetic compounds alter neuron motor proteins will be important in defining how anesthetics alter neuronal signaling and energetic dynamics.
Citation: Dubey P, Datta R, Eckenhoff RG, Bedell VM (2026) In vivo changes in zebrafish anesthetic sensitivity in response to the loss of kif5Aa are associated with the alteration of mitochondrial motility. PLoS One 21(7): e0316959. https://doi.org/10.1371/journal.pone.0316959
Editor: Miriam A. Hickey, University of Tartu, ESTONIA
Received: December 18, 2024; Accepted: May 6, 2026; Published: July 27, 2026
Copyright: © 2026 Dubey et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Data is available at https://doi.org/10.5061/dryad.x3ffbg808.
Funding: University of Pennsylvania, Department of Anesthesiology, provided salary support for VMB and PD. Foundation for Anesthesiology Education and Research’s Mentored Research Training Grant to [VMB] provided financial support to purchase lab supplies. The National Institute of General Medical Sciences [RM1 GM13651] to [VMB] provided financial support to purchase lab supplies and partial coverage of VMB salary. Center for Undergraduate Research and Fellowships at the University of Pennsylvania to [RD] provided salary coverage for the summer. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
There is a remarkable variety of small compounds capable of producing changes in consciousness, ranging from sedation to general anesthesia. These drugs have many incompletely understood effects on the central nervous system (CNS), including energy production and use [1,2]. The neuronal synapse has one of the highest energy requirements in the brain [3,4] and the highest mitochondrial concentration [5]. Additionally, mitochondria are targeted by volatile anesthetics [6] by specifically inhibiting complex I of the electron transport chain and decreasing synaptic ATP [7]. Decreased synaptic excitatory transmission is observed in animal models with anesthetic hypersensitivity caused by cell-specific impaired mitochondrial function [6,7]. However, action upon the electron transport chain does not fully define anesthetics’ effect on cellular energy availability.
Complex systems have evolved to transport neuronal mitochondria from their site of biogenesis in the cell body [8] to and from the synapse [9], presumably to align with synaptic energy needs [10]. Kinesin (kif) is the motor protein family responsible for the anterograde movement of various cargoes within cells [11]. While over 80% of the kinesin superfamily are expressed in neurons [11], three kinesin families, Kif5, Kif1, Kinesin like protein 6 (Klp6) have been shown to transport mitochondria within the axons [12–14].
Of these kinesins, previous studies on propofol-binding partners identified kif5B as a target [15]. Propofol binds with high affinity to the motor head domain of kif5B, causing premature cargo detachment from the microtubule [15–17]. Additionally, the propofol binding pocket within kif5B has been identified as requiring two specific amino acids within the motor domain (S1 Fig). While the motor domain is largely conserved, it is still unknown if the intravenous (IV) anesthetics bind to other kinesins. Additionally, it remains unknown if changing kinesin motility alters behavioral outcomes in vivo. Therefore, to answer these questions, we employed the zebrafish model organism.
With the zebrafish, kif5 family expression patterns are known [18]. kif5Ba and kif5Bb are ubiquitously expressed throughout the developing animal. However, kif5Aa is primarily expressed within the nervous system and is a key mitochondrial transport protein [18–20]. Further, we found significant homology within the motor domain and propofol binding pocket between Kif5Aa and Kif5B (S1 Fig). Therefore, we decided to employ the kif5Aa knockout (KO) zebrafish line as an intersection between motor domain homology and mitochondrial motility.
The kif5Aa KO line has a point mutation that alters a critical splice site, resulting in a premature stop codon and the truncation of the native protein, eliminating the cargo binding domain [19]. Previous studies have shown that homozygous kif5Aa KO larvae exhibit hyperexcitability, which predisposes them to seizures, [19] and a loss of response to visual stimuli [19,20]. Additionally, the larvae develop peripheral neuropathy and nerve degeneration starting at 6–8 days post fertilization (dpf) [19]. These defects, along with a poorly developed swim bladder, are lethal between 8–12 dpf. How the loss of kif5Aa affects larval movement and central nervous system axonal mitochondrial movement is still an open question.
Here, we utilize kif5Aa KO larvae at 5 dpf, prior to the onset of the neural degeneration phenotype, to assess whether sensitivity to four distinct IV anesthetic/sedative drugs is altered using whole-animal behavioral assays. Additionally, we assess for changes in mitochondrial motility using primary neuronal cell culture.
Methods
Zebrafish husbandry
All zebrafish experiments complied with the University of Pennsylvania Institutional Animal Care and Use Committee (IACUC). The adult zebrafish were maintained in an aquatic facility overseen by the University Laboratory Animal Resources (ULAR) at the University of Pennsylvania.
All zebrafish were maintained at 28 °C with a 13−11 light-dark cycle as previously described [21]. Adult fish were maintained using standard husbandry conditions [21]. Knockout (KO) and wild type (WT) siblings were maintained in the same conditions until experimentation.
Genotyping the kif5Aa KO line
Genomic DNA was obtained from adult animals using the fin-clipping method [21]. The PCR identifying the kif5Aasa7168 KO animals was run using previously described primers and digested using Dra III [19].
kif5Aasa7168 heterozygous zebrafish were mated, and before behavioural testing, the larvae were separated based on known visual phenotypes: loss of the swim bladder and increased pigmentation. “WT” were a combination of WT and heterozygous animals as they were indistinguishable.
The kif5Aa heterozygous line was outcrossed into a stable Tg(actb2:mito Grx1-roGFP2) kindly given to us by Michael Pack’s lab. This construct was created along with the cytosolic version [22] using the mito Grx1-roGFP2 construct (Addgene plasmid 64977) obtained from Tobias Dick [23]. A fragment of the construct was cloned behind the zebrafish actb2 promoter (using p5E-bactin2) [24] with the multisite gateway system as previously described, then integrated into the zebrafish genome with Tol2 transposase [24,25]. mito Grx1-roGFP2 [23] targets the roGFP2 to the mitochondria using the signal sequence from Neurospora crassa ATP synthase protein fused to human Grx1/GLRX followed by roGFP [23]. To align with the previously published annotation and to avoid a rather lengthy name (Tg(actb2:Ncrassa.ATP5G3 MTS- Hsa.GLRX-roGFP2)) we refer to the transgenic line here as Tg(actb2:mito Grx1-roGFP2).
Drug exposures
We tested four sedative/anesthetics: propofol, etomidate, dexmedetomidine, and ketamine. Each drug was diluted into E3 embryo water for the behavior experiments described previously [26]. Briefly, propofol was diluted to 100 mM from a 100 mM DMSO stock in E3 by sonicating 5 minutes (min); absorbance at 270nm was used to confirm the concentration. Etomidate and dexmedetomidine were solubilized directly by vortexing. Finally, ketamine was diluted at in E3 with 5mM hepes at a pH of 8, which is required for improved uptake by the larvae.
For cell culture, all drugs were initially diluted to a 2x concentration using 1x PBS. This was then diluted to the final drug concentration in complete cell culture media to ensure adequate nutrition for the cells.
Zebrafish 5 dpf behavioural assay
Behavior experiments were performed as previously described [26]. Briefly, we used the DanioVision behavioral system with EthoVision software (Noldus) to measure the distance the WT and kif5Aa KO larvae moved spontaneously (SM) over 4 min or in 1 second after an elicited acoustic stimulus (EM). The experiments were 2 hours within the dark enclosure of the behaviour system, with the EM at the highest setting every 30 minutes. All experiments used 96-well glass plates (Chemglass Life Sciences, CG-1910) at 28oC.
Primary zebrafish neuronal cell culture
This protocol is based on a previously described zebrafish neuronal cell culture assay [27]. Briefly, we used a 96-well tissue culture plate with a glass-like polymer coverslip bottom (Cellvis, P96-1.5P) coated with Entactin-Collagen-Laminin (EMD Millipore, 08–110) at 10–20 μg/mL and stored at 4 °C overnight.
Twenty 5 dpf zebrafish WT siblings and kif5Aa KO larvae containing the Grx ro-GFP2 were euthanized on ice until the heart stopped beating. Brains were dissected and placed in 500 μL Hanks base salt solution with 1% penicillin/streptomycin on ice. 200 μL of Accutase (Sigma, A6964) was added for dissociation and incubated for 10 min in a 37 °C water bath. Brains were triturated using a glass pipette for 2–4 min or until there were few visible clumps of cells. This was key to decreasing aggregation and promoting attachment and axon outgrowth during cell culturing. The dissociation reaction was stopped using 700 μL of complete media. Complete media consisted of 0.22 μm sterile-filtered Leibovitz-15, 2.5mM Glutamax I, 15 ng/mL epidermal growth factor, 10% fetal bovine serum, 1% penicillin/streptomycin and 5% zebrafish embryo extract. The embryo extract was created as previously described [28].
The sample was centrifuged at 1500 RPM for 6 min at 4 °C. The supernatant was discarded and 100–125 μL of complete media was added to resuspend the pellet. The cell mixture was plated using 50–75 μL per well. The cells were incubated at 28 °C for 2 days.
Imaging
Neurons were imaged using a Leica DMI4000 with a Yokagawa CSU-X1 spinning disk confocal attachment at 63x.
The anaesthetic concentration chosen was the EC50s previously determined [26]: propofol 3 μM, etomidate 30 μM, dexmedetomidine 10 nM, and ketamine 130 μM. The cells were incubated in the drug for at least 30 min before imaging. Neurons that could be traced back to a specific cell body were imaged every 3 seconds for 5 minutes for a total of 101 images (initial image plus imaging for the next five minutes). Five to ten cells were imaged per well.
An open-source ImageJ plugin, KymoToolBox, defined mitochondrial motility. S1 Table is a summary of the mitochondria traced. Briefly, each movie contained a range of 1–6 axons. The axons were highlighted, and a kymograph was created, on which individual mitochondria were traced. The number of traceable mitochondria varied between movies, therefore, the range of mitochondria traced is given in S1 Table. Any movie with less than 5 traceable mitochondria was discarded. ImageJ was then able to calculate the distance moved for the mitochondria. Total mitochondrial movement is herein defined as any movement, retrograde plus anterograde, including zero movement. No minimum motility down the axon was used.
Statistics
For behavior studies, the distance moved for 8–12 larvae per drug concentration per genotype was averaged and counted as a single experimental value (n). The experiment was performed at least 4 times on separate days to ensure technical and biological replication. Each animal was exposed to a single drug and not reused. For the baseline movement assays, 40–41 experiments at 30 minutes were analyzed and significance tested using the Mann-Whitney test for nonparametric data. Hill curves were created using 6–8 drug concentrations, including a no-drug (ND) control. The distance moved was normalized using the ND as 100% movement. Hill curves were created using GraphPad Prism. As with prior studies, the Hill curves were constrained to a top value of 100% and a bottom of 0%. The EC50s with 95% confidence intervals (CI) were calculated and plotted. EC50 comparison p values were determined using a sum of squares F test.
All experiments were performed at least in triplicate for the mitochondrial motility assays. The motility data was, in most cases, normally distributed. Therefore, a two-way ANOVA using Bonferroni’s correction for multiple comparisons determined significance. For all groups, significance is labeled: not significant (n.s., p > 0.05), p ≤ 0.05 (*), p ≤ 0.01 (**), p ≤ 0.001 (***), p ≤ 0.0001.
Results
Loss of kif5Aa alters behavioral sensitivity in a drug-specific manner
We examined larval movement using two previously described behavioral endpoints: loss of spontaneous movement (SM) and loss of elicited movement (EM) to an acoustic tap stimulus [26]. Initially, we tested baseline movement to the two behaviors at 5 dpf. The kif5Aa KO larvae significantly increased SM compared to their wild-type siblings (WT) (S2A Fig). In contrast, the kif5Aa KO animals showed significantly decreased EM (S2B Fig). For all subsequent studies, movement of no drug controls was defined as 100%.
Because it was unclear if the depletion of synaptic mitochondria was a time-dependent phenomenon during the maintenance phase of anesthesia, we decided to test for changes in sensitivity over time. First, we assessed in vivo sensitivity to propofol over time in the 5 dpf larvae. Using SM, the kif5Aa KO was more sensitive to propofol than their WT siblings (Fig 1A–1D), with a 30–40% decrease in the EC50 (Fig 1E). Using EM, there was no initial difference in sensitivity (Fig 1F–1J), but after 90 minutes of exposure, the kif5Aa KO became significantly more sensitive to propofol than WT (Fig 1H–1I). We were able to see a decrease in the EC50 of the kif5Aa KO larvae over time (Fig 1J).
A-D) Increased sensitivity to propofol in the kif5Aa KO larvae (red) as compared to WT (blue) using spontaneous movement (SM) is consistent over time. E) A significant decrease in the EC50 of kif5Aa KO versus WT over time. F-G) No significant change in sensitivity was seen with elicited movement. H-I) By 90 minutes, there was a significant sensitivity to propofol with EM. J) The EC50s within kif5Aa KO larvae decreased over time and showed sensitivity to propofol at 90- and 120-min. Hill curves: 7 propofol concentrations (0-30 mM), n = 7-9 (average of 8-12 zebrafish per genotype per drug dose), SEM error bars. EC50s: calculated from Hill curves, statistical significance using the extra sum of squares F test, 95% CI error bars.
Next, we assessed sensitivity to etomidate and found kif5Aa KO larvae were 85–90% more sensitive using the SM endpoint compared to the WT siblings (Fig 2A–2E). When using the EM endpoint, like propofol, when exposed to etomidate the kif5Aa KO larvae had no significant sensitivity at 30–60 minutes (Fig 2F–2G). However, by 90 minutes kif5Aa KO larvae were significantly more sensitive to etomidate (Fig 2H–2I). The larvae demonstrated a slow, time-dependent increase in sensitivity to etomidate (Fig 2J).
A-D) When testing spontaneous movement (SM), the kif5Aa KO larvae (red) were more sensitive to etomidate as compared to their WT siblings (blue). This sensitivity was consistent over time. E) There was a significant decrease in the EC50 of kif5Aa KO compared to WT siblings, which is consistent over time. F-G) Elicited movement (EM) showed no significant difference in Kif5Aa KO compared to WT. H-I) At 90 and 120 minutes, there were significant increase in sensitivity in kif5Aa KO larvae vs WT siblings. J) Time-dependent decrease in EC50s in kif5Aa KO larvae, which became significant at 90 and 120 minutes. Hill curves: 9 etomidate concentrations (0 to 100 mM), n = 7-13 (average of 8-12 larval zebrafish per n), SEM. EC50s: calculated from the Hill curves, statistical significance using the extra sum of squares F test, 95% CI error bars.
Given the consistent SM behavioral sensitivity, we examined a sedative drug without full anesthetic properties, dexmedetomidine. We found a time-dependent resistance to dexmedetomidine in the kif5Aa KO larvae (Fig 3A–3C) that resolved by 120 minutes (Fig 3D). We found a 2–4-fold increase in the EC50 for the first 90 minutes (Fig 3E). Consistent with dexmedetomidine being only sedative, there was no loss of movement to the EM (Fig 3F–3I).
A-D) When testing spontaneous movement (SM), kif5Aa KO larvae (red) were found to be resistant to dexmedetomidine as compared to their WT siblings (blue). D) This resistance resolved over time. E) The EC50s showed a significant increase in the kif5Aa KO larvae as compared to WT, which resolved by 120 min. H-I) Dexmedetomidine was a known sedative that did not induce general anesthesia. There was no loss of movement to the EM. At supratherapeutic doses of dexmedetomidine (0.1 and 0.3 mM, last two points on the blue line), there was an increase in WT movement that was not seen in the kif5Aa KO larvae. Hill curves: 7 dexmedetomidine concentrations (0 to 0.3 mM), SEM error bars. EC50s: n = 5-8, (average of 8-12 larval zebrafish per genotype per drug dose). EC50s: calculated from the Hill curves, statistical significance using the extra sum of squares F test, 95% CI error bars.
Ketamine, a non-GABAergic anaesthetic, produced no change in the loss of SM (Fig 4A–4D) and the EC50s of WT to kif5Aa KO larvae overlapped (Fig 4E). Similarly, there was no change in response to EM when exposed to ketamine (Fig 4F–4J).
A-D) When testing SM, the Hill curve does not change in response to ketamine. E) The EC50s of WT (blue) and kif5Aa KO (red) overlap. F-I) Similarly, there was no change in EM in response to ketamine. J) The EC50s of kif5Aa KO (red) and their WT siblings (blue) are equivalent. Due to the lack of phenotype, it was decided to preserve animals and not further replicate the data to improve the 95% CI of the EC50s.
Mitochondrial motility is altered in a drug-specific manner
We next asked whether alterations in anesthetic sensitivity were associated with changes in mitochondrial motility within primary neuron cells cultured from both WT (S3A Fig) and kif5Aa KO (S3B Fig) brains. All mitochondria within the neuron were labeled using a stable Tg(actb2:mito Grx1-roGFP2) transgenic line showing the interconnected network the organelle creates. Time-lapse moves were created of the neurons, and the axons were highlighted in Image J (S4A-S4A’ Fig, S4B-S4B’ Fig, S1 Movie). Using these movies, kymographs were created, and the mitochondria were traced over time (S4A”, S4B” Fig).
Baseline mitochondrial motility with no drug (ND) was significantly different in the kif5Aa KO neurons as compared to the WT for every measurement, including motility, distance moved, and velocity. Further, this difference was seen in total movement, anterograde and retrograde (first ND blue bar and first ND red bar, Figs 5-8).
Mitochondrial motility was assessed in cultured primary zebrafish neurons. When comparing no drug (ND) WT to kif5Aa KO mitochondrial movement, there was a decrease in the percent of motile mitochondrial, distance moved, and velocity despite directionality (A-I, first red bar and first blue bar). A) There was a decrease in the percent of motile mitochondria in WT neurons when exposed to propofol (blue bars). B-C) There was no change in the total distance moved or velocity (blue bars) within WT neurons. D-F) When assessing anterograde motility in WT neurons, there was a significant decrease in the number of motile mitochondria, the distance moved, and velocity (blue bars). G) In WT neurons, there was a decrease in the number of motile retrograde mitochondria (blue bars), (H) no change in the distance moved (blue bars) but a (I) decrease in retrograde velocity (blue bars). A-I) When kif5Aa KO neurons were exposed to propofol, there was no significant change in movement (red bars). n = 31-68. Error was in standard deviation and statistical significance using a two-way ANOVA with multiple comparisons using Bonferroni’s correction.
Total motility (A-C) was assessed and then broken into anterograde (D-F) and retrograde (G-I) movement. A) There was a significant decrease in motile mitochondria within the WT cells (blue bars) as well as a decrease in total distance moved (B, blue bars) and velocity (C, blue bars). When assessing anterograde movement, there was a decrease in the number of motile mitochondria (D, blue bars), the distance moved (E, blue bars), and the velocity (F, blue bars). Additionally, when assessing retrograde motility, we found a decrease in the number of motile mitochondria (G, blue bars), distance moved (H, blue bars), and velocity (I, blue bars). A-I) However, within the kif5Aa KO neurons, there was no significant change in motility when exposed to etomidate (red bars). n = 25-68 (see Supplemental Table 2). Error was in standard deviation and statistical significance using a two-way ANOVA with multiple comparisons using Bonferroni’s correction.
A-I) There was no significant change in mitochondrial motility (blue bars) when WT neurons were assessed. There was a significant increase in motile mitochondria within the kif5Aa KO cells (A, red bars) and in the total distance moved (B, red bars). However, the velocity (C, red bars) was not significantly different. When assessing anterograde motility, there was an increase in the number of motile mitochondria (D, red bars) and their velocity (F, red bars) but not in the distance they moved (E, red bars). Retrograde mitochondrial motility showed a similar number of motile mitochondria (G, red bars) but an increase in the distance moved (H, red bars) and velocity (I, red bars). When the immobile mitochondria were removed for the average, we assessed WT neuron mitochondria motility. We found a significant increase in the WT mitochondrial retrograde distance traveled (J, blue bars) and velocity (K, blue bars) but not kif5Aa KO mitochondria (J-K, red bars). n = 30-68. Error was in standard deviation and statistical significance using a two-way ANOVA with multiple comparisons using Bonferroni’s correction.
A-I) Within WT neurons, there was no significant change in mitochondrial motility in response to ketamine (blue bars). Within the kif5Aa KO neurons, there was an increase in the total number of motile mitochondria (A, red bars), with that increase being primarily anterograde movement (D, red bars) and not retrograde (G, red bars). B-D, red bars) There was no change in the total distance moved or velocity. There was an increase in anterograde velocity (F, red bars) but no change in the distance moved (E, red bars). G-I) There was no change in retrograde mitochondrial movement within the kif5Aa neurons. n = 30-68.Error was in standard deviation and statistical significance using a two-way ANOVA with multiple comparisons using Bonferroni’s correction.
In WT neurons, propofol produced a significant decrease in the total number of motile mitochondria (Fig 5A, blue bars) without a change in the total distance moved (Fig 5B, blue bars) or in the total velocity (Fig 5C, blue bars). Dividing the total movement by direction, propofol produced a significant decrease in anterograde mitochondrial motility, distance moved, and velocity (Fig 5D–5F, blue bars). There was also a decrease in retrograde motility and velocity (Fig 5G, 5I, blue bars) but no change in retrograde distance (Fig 5H, blue bars). In contrast, propofol produced no motility changes in kif5Aa KO neurons (Fig 5A–5I, red bars).
Etomidate significantly reduced the number of motile mitochondria (Fig 6A, blue bars), the total distance mitochondria moved (Fig 6B, blue bars), and the mitochondrial velocity (Fig 6C, blue bars) in WT neurons. Dividing into anterograde and retrograde directions, etomidate produced a significant decrease in the number of motile mitochondria, distance moved, and velocity in both directions (Fig 6E–6I, blue bars). Etomidate produced no change in the kif5Aa KO neuron mitochondrial motility (Fig 6A–6I,red bars).
In contrast, dexmedetomidine produced no significant changes in mitochondrial motility or velocity from WT neurons (Fig 7A–7I, blue bars). However, a trend existed for increased distance moved and velocity in the retrograde direction (Fig 7H–7I, blue bars). If we focused on the motile mitochondria (56% of total), dexmedetomidine increased the distance moved and velocity (Fig 7J–7K).
Within the kif5Aa KO neurons, dexmedetomidine exposure produced a significant increase in the total number of motile mitochondria (Fig 7A, red bars), and distance moved (Fig 7B, red bars), without changing velocity (Fig 7C, red bars). Dividing motility into anterograde and retrograde in the kif5Aa KO neurons, dexmedetomidine caused significantly more mitochondria to move anterograde (Fig 7D, red bars), coupled with an increase in velocity (Fig 7F, red bars), but without a significant change in the distance traveled (Fig 7E, red bars). Dexmedetomidine also caused a significant increase in both the retrograde distance moved (Fig 7H, red bars) and the velocity (Fig 7I, red bars) of the mitochondria. When focusing on the motile mitochondria, we saw no significant increase in either distance traveled (Fig 7J, red bars) or velocity (Fig 7K, red bars).
Similar to dexmedetomidine, ketamine produced no significant alteration in the motility of WT neuronal mitochondria (Fig 8A–8I, blue bars), although there was an increase in the number of motile mitochondria within the kif5Aa KO neurons (Fig 8A). However, this did not lead to any change in distance moved or velocity (Fig 8B–8C). In the kif5Aa KO neurons, ketamine increased anterograde mitochondrial motility and velocity (Fig 8D, 8F) but no change was detected in the distance moved (Fig 8E). Finally, ketamine produced no change in retrograde mitochondrial motility in the kif5Aa KO neurons (Fig 8G–8I).
Discussion
Here we use the zebrafish model organism to expand upon the effect various IV anaesthetic/sedative drugs have upon intracellular motors, specifically Kif5A, as reflected by mitochondrial motility. When comparing human Kif5B with zebrafish Kif5Aa, there is a 67% overall homology, with very high homology in the motor domain (S1 Fig, blue line). Additionally, residues defining the previously published propofol binding site [29] are conserved (S1 Fig, red boxes). However, previous studies were performed in silico, in vitro, or in cell culture [15–17,29]. Therefore, it was unknown if propofol binding to Kif5A was associated with altered behavioural responses to sedatives/anaesthetics. Additionally, since a known cargo of Kif5A is mitochondria, it was unclear whether IV drugs altered their distribution in the neuron, with potentially important energetic consequences.
Here, we provide data suggesting that mitochondrial transport is important in the behavioral sensitivity to certain sedation/anesthetic drugs. Our behavioral data showed that the kif5Aa KO larvae were significantly more sensitive to the GABAeric drugs propofol and etomidate compared to their WT siblings and that sensitivity to etomidate was greater than to propofol. This sensitivity within the mutant animals was consistent with the in vitro data. It had been shown that, for propofol, the motor domain of kinesin was bound, which induced the motor protein to fall off the microtubule. This, theoretically, caused a decrease in energy available at the synapse. However, within the kif5Aa KO larvae, there is a decrease in anterograde movement and, theoretically, a dearth of mitochondria at the synapse. Therefore, the effect of this binding was accounted for prior to induction. Thus, the energy-poor synapses would be more vulnerable to the other mechanisms of the anesthetics.
Unexpectedly, the KO animals’ sensitivity to the other drugs, ketamine and dexmedetomidine, was less evident; in fact, we noted a decrease in sensitivity to the alpha 2 agonist dexmedetomidine. The complete lack of a behavioral phenotype to ketamine was unanticipated. However, it is still unknown if the kinesin motor domain is bound and inhibited by ketamine. Alternatively, the kif5Aa KO larvae were resistant to dexmedetomidine. This suggests other protein binding partners within the motor protein pathway for these compounds.
When considering neuron energy homeostasis, previous studies within mice on the loss of a critical protein in the electron transport complex 1 (ndufs4), caused an increase in sensitivity to propofol and a resistance to ketamine [30]. Therefore, if the kif5Aa motor protein phenotype was only due to energetics, this data would be inconsistent. However, unlike the ndufs4 protein, altering motor protein transport also altered mitochondrial health and influenced fission/fusion events within the neuron. Adding to the complexity, mitochondria are not the only anaesthetic-bound protein/organelle to be transported by the kif5A protein family. It has been demonstrated that kif5A is crucial in the transport of the GABAA receptor [31]. Therefore, it is likely that mitochondria are not the only cargo that influences the anesthetic phenotype. Likely, the alteration of neuronal subcellular processes is complex and compound-specific.
Our in vivo studies used two fundamentally different behavioral endpoints to study anesthetic/sedative compounds. These two endpoints responded differently over time. Within the GABAergic drugs, propofol and etomidate, SM was consistent, whereas EM had a time-dependent phenotype. SM was a more complex behavior that involved decision-making circuits within the larvae [32]. However, given the complexity, it would theoretically be easier for a drug to disrupt those neural circuits and thus may explain why we see a consistent phenotype over time. Alternatively, the neural circuits regulating EM has been elucidated. This behavior was a less complex but evolutionarily critical reflex arc [33–35]. Theoretically, these critical reflex neurons would require more time to deplete their energy. Therefore, we see a slow increase in sensitivity over time.
Consistent with the in vitro single kinesin motor protein and binding studies [15,16], mitochondria within WT neurons moved less when exposed to both GABAergic drugs. Propofol produced anterograde movement inhibition, whereas etomidate inhibited both anterograde and retrograde movement. These results suggest an additional effect on dynein and/or its accessory proteins. Consistent with dexmedetomidine and ketamine acting on different molecular targets than etomidate or propofol, they had a very different effect on mitochondrial transport. Ketamine had little effect, while dexmedetomidine enhanced mitochondrial motility, especially in the retrograde direction.
Knocking out kif5Aa reduced mitochondrial anterograde transport, and the incremental effect of adding the GABAergic anesthetics was eliminated. This suggests that Kif5Aa may be the key anterograde neuronal motor protein targeted by those drugs. Despite the decrease in motility, anterograde transport is still seen within the KO cells. Therefore, other kinesin proteins may compensate for the loss of the single kinesin but incompletely. For example, while Kif5Aa appears to be the important motor protein for neuron mitochondria, there is a similar protein, Kif5Ab. Therefore, in kif5Aa KO fish, kif5Ab may be upregulated. Alternatively, kif1 and klp6 families also contribute to neuron mitochondrial anterograde motility [12–14]. Nevertheless, despite neurons containing kinesins that may partially compensate for the loss of Kif5Aa, when the kif5Aa KO neurons were exposed to propofol and etomidate, there was no further decrease in mitochondrial motility. This suggests that neither propofol nor etomidate targets these alternative kinesins.
Retrograde motility is dependent upon dynein motor proteins [36,37]. To our knowledge, the effect of anesthetics/sedatives on this key neuronal regulatory protein has not been studied. Here, we show that etomidate and dexmedetomidine influence retrograde motility. With etomidate, there is a decrease in WT mitochondria retrograde motility. Additionally, although not statistically significant, a trend toward decreased motility was observed in kif5Aa KO neurons, which is not seen in propofol, suggesting that a subset of neurons may be affected by etomidate. Compared to etomidate, dexmedetomidine does not alter the number of motile retrograde mitochondria within WT cells but does increase their distance and speed. Additionally, there is an increase in mitochondrial retrograde motility within the kif5Aa KO cells. This suggests that dexmedetomidine enhances mitochondrial transport from the synapse to the soma. It is still unclear if this alters synaptic mitochondrial health.
When interpreting these data, there are two key considerations. First, these data were collected from neurons cultured from whole brains, which include many different neuronal populations: motor, sensory, inhibitory, and excitatory. Given this heterogeneity, it is unsurprising that we observe large variability when assessing neuron mitochondrial motility. It is still unknown if the mitochondrial motility phenomenon is pan-neural or if different neurons are differentially affected. Additionally, it remains unknown whether the loss of kif5Aa causes specific subtypes of neurons to either fail to develop or undergo apoptosis. This may cause a difference in the neuron cell populations within the cell culture. Therefore, the next step is to fully utilize the power of the zebrafish model by obtaining animals with fluorescently labeled neuronal subsets and assessing motility within those various classes.
Second, it is important to consider that the nervous system of 5 dpf fish is still growing and making critical connections. Therefore, the brain may respond differently to changes in motor and energetic homeostasis compared to the adult brain. It is still unknown if the mechanisms described above will continue into adulthood.
In conclusion, anesthetic/sedative drugs influence whole animal behavior and mitochondrial motility similarly. Using a kif5Aa knock-out animal has enabled the conclusion that this class of motor protein produces changes in mitochondrial motility, and some anesthetic effects may be translated through these proteins. These data suggest the hypothesis that alteration of mitochondrial dynamics and distribution contribute to the anesthetic state.
Supporting information
S1 Table. A summary of the number of mitochondria traced for the neuronal motility assays.
We assessed two genotypes, WT and kif5Aa KO neurons, and exposed them to 4 drugs with a no-drug (ND) control. The ND neurons were imaged in every experiment; hence the numbers were greater than that of the exposure groups. The “n” used and plotted as an open circle in Figs 4–6 and Supplemental Fig 6, was the total number of movies created. Within each movie, there was a range of traceable mitochondria, with a minimum allowable number of 5. The final column showed the total number of mitochondria traced per genotype per exposure group.
https://doi.org/10.1371/journal.pone.0316959.s001
(PDF)
S1 Fig. Zebrafish kif5Aa and human kif5B demonstrates homology within the motor domain.
There is 67% homology between the paralogs Zebrafish kif5Aa and human kif5B. In the motor domain, the first approximately 340 amino acids, showed high sequence conservation (blue line). Additionally, the previously described amino acids directly binding propofol, leucine 258 and phenylalanine 318 (red boxes), were conserved.
https://doi.org/10.1371/journal.pone.0316959.s002
(PDF)
S2 Fig. Baseline movement was different in WT siblings versus the kif5Aa KO larvae.
Movement of the 5 dpf WT sibling and kif5Aa KO larvae was assessed using the two behavioral assays, spontaneous (SM) and elicited movement (EM). A) The WT siblings demonstrated significantly less SM as compared to the kif5Aa KO larvae despite the loss of a swim bladder to maintain buoyancy. B) In contrast, when assessing EM, an acoustic startle response, there was a significant decrease in movement. The data was collected from the no-drug controls across all experiments using the 30-minute time point. The n = 40–41 with each point consisting of the average movement of 8–12 larvae. Error bars use standard deviation. Statistical significance using the Mann-Whitney test for nonparametric data.
https://doi.org/10.1371/journal.pone.0316959.s003
(PDF)
S3 Fig. Examples of WT and kif5Aa KO primary zebrafish neurons.
The whole brains of kif5Aa KO larvae (B) and their WT siblings (A) were used to create primary cell cultures. A-B) Example of a bright field plate at 20x magnification. Due to the whole brain being used for cell culture, most of the neurons were surrounded by other nonneuronal cells. A’-A”’) Three example WT neurons overlaying brightfield and GFP showed the neuron cell boundary and the mitochondria within the cell, respectively. B’-B”’) Three example kif5Aa KO neurons, which demonstrated similar morphology to that of the WT siblings.
https://doi.org/10.1371/journal.pone.0316959.s004
(PDF)
S4 Fig. WT and kif5Aa KO primary zebrafish neurons were used to measure mitochondrial motility.
The whole brains of kif5Aa KO larvae (B) and their WT siblings (A) were used to create primary cell cultures. A’-B’) The neurons were imaged every 3 seconds for 5 minutes, and the axons were traced in ImageJ. The movies of these axons were used to create kymographs. A”-B”) Kymographs were used to trace the mitochondria. Black is the background, and white shows mitochondrial fluorescence. Blue (A”) and red (B”) were a single example of tranced mitochondria.
https://doi.org/10.1371/journal.pone.0316959.s005
(PDF)
S1 Movie. Example WT and kif5Aa KO axon.
An example axon from WT (top) and kif5Aa KO (bottom) neurons. Mitochondria can be seen moving both anterograde and retrograde within these axons. This movie demonstrates the difference between baseline motility in WT versus KO axons.
https://doi.org/10.1371/journal.pone.0316959.s006
(MOV)
Acknowledgments
Dr. Florence Marlow (Icahn School of Medicine at Mount Sinai) kindly provided the kif5Aasa7168 KO line. Drs. Philip Morgan and Margaret Sedensky for their input on the manuscript.
Declaration of generative AI and AI-assisted technologies in the writing process: During the preparation of this work, the author(s) used Grammarly to improve the document’s grammar. After using this tool/service, the author(s) reviewed and edited the content as needed and took (s) full responsibility for the content of the publication.
References
- 1. Campagna JA, Miller KW, Forman SA. Mechanisms of actions of inhaled anesthetics. N Engl J Med. 2003;348(21):2110–24. pmid:12761368
- 2.
Hemmings HC Jr, Herold KF. Mechanisms of Intravenous Anesthetic Action. Total Intravenous Anesthesia and Target Controlled Infusions. Springer International Publishing. 2017. 79–95. https://doi.org/10.1007/978-3-319-47609-4_5
- 3. Harris JJ, Attwell D. The energetics of CNS white matter. J Neurosci. 2012;32(1):356–71. pmid:22219296
- 4. Harris JJ, Jolivet R, Attwell D. Synaptic energy use and supply. Neuron. 2012;75(5):762–77. pmid:22958818
- 5. Wong-Riley MT. Cytochrome oxidase: an endogenous metabolic marker for neuronal activity. Trends Neurosci. 1989;12(3):94–101. pmid:2469224
- 6. Zimin PI, Woods CB, Quintana A, Ramirez J-M, Morgan PG, Sedensky MM. Glutamatergic Neurotransmission Links Sensitivity to Volatile Anesthetics with Mitochondrial Function. Curr Biol. 2016;26(16):2194–201. pmid:27498564
- 7. Zimin PI, Woods CB, Kayser EB, Ramirez JM, Morgan PG, Sedensky MM. Isoflurane disrupts excitatory neurotransmitter dynamics via inhibition of mitochondrial complex I. Br J Anaesth. 2018;120(5):1019–32. pmid:29661379
- 8. Davis AF, Clayton DA. In situ localization of mitochondrial DNA replication in intact mammalian cells. J Cell Biol. 1996;135(4):883–93. pmid:8922374
- 9. Devine MJ, Kittler JT. Mitochondria at the neuronal presynapse in health and disease. Nat Rev Neurosci. 2018;19(2):63–80. pmid:29348666
- 10. Li S, Sheng Z-H. Energy matters: presynaptic metabolism and the maintenance of synaptic transmission. Nat Rev Neurosci. 2022;23(1):4–22. pmid:34782781
- 11. Hirokawa N, Noda Y, Tanaka Y, Niwa S. Kinesin superfamily motor proteins and intracellular transport. Nat Rev Mol Cell Biol. 2009;10(10):682–96. pmid:19773780
- 12. Lin M-Y, Sheng Z-H. Regulation of mitochondrial transport in neurons. Exp Cell Res. 2015;334(1):35–44. pmid:25612908
- 13. Nangaku M, Sato-Yoshitake R, Okada Y, Noda Y, Takemura R, Yamazaki H, et al. KIF1B, a novel microtubule plus end-directed monomeric motor protein for transport of mitochondria. Cell. 1994;79(7):1209–20. pmid:7528108
- 14. Wozniak MJ, Melzer M, Dorner C, Haring H-U, Lammers R. The novel protein KBP regulates mitochondria localization by interaction with a kinesin-like protein. BMC Cell Biol. 2005;6:35. pmid:16225668
- 15. Bensel BM, Guzik-Lendrum S, Masucci EM, Woll KA, Eckenhoff RG, Gilbert SP. Common general anesthetic propofol impairs kinesin processivity. Proc Natl Acad Sci U S A. 2017;114(21):E4281–7. pmid:28484025
- 16. Woll KA, Guzik-Lendrum S, Bensel BM, Bhanu NV, Dailey WP, Garcia BA, et al. An allosteric propofol-binding site in kinesin disrupts kinesin-mediated processive movement on microtubules. J Biol Chem. 2018;293(29):11283–95. pmid:29844014
- 17. Frank M, Nabb AT, Gilbert SP, Bentley M. Propofol attenuates kinesin-mediated axonal vesicle transport and fusion. Mol Biol Cell. 2022;33(13):ar119. pmid:36103253
- 18. Campbell PD, Marlow FL. Temporal and tissue specific gene expression patterns of the zebrafish kinesin-1 heavy chain family, kif5s, during development. Gene Expr Patterns. 2013;13(7):271–9. pmid:23684767
- 19. Campbell PD, Shen K, Sapio MR, Glenn TD, Talbot WS, Marlow FL. Unique function of Kinesin Kif5A in localization of mitochondria in axons. J Neurosci. 2014;34(44):14717–32. pmid:25355224
- 20. Auer TO, Xiao T, Bercier V, Gebhardt C, Duroure K, Concordet JP, et al. Deletion of a kinesin I motor unmasks a mechanism of homeostatic branching control by neurotrophin-3. Elife. 2015;4. pmid:26076409
- 21.
Westerfield M. The zebrafish book: A guide for the laboratory use of zebrafish (Danio rerio). 4 ed. University of Oregon Press. 2000.
- 22. Seiler C, Davuluri G, Abrams J, Byfield FJ, Janmey PA, Pack M. Smooth muscle tension induces invasive remodeling of the zebrafish intestine. PLoS Biol. 2012;10(9):e1001386. pmid:22973180
- 23. Gutscher M, Pauleau A-L, Marty L, Brach T, Wabnitz GH, Samstag Y, et al. Real-time imaging of the intracellular glutathione redox potential. Nat Methods. 2008;5(6):553–9. pmid:18469822
- 24. Kwan KM, Fujimoto E, Grabher C, Mangum BD, Hardy ME, Campbell DS, et al. The Tol2kit: a multisite gateway-based construction kit for Tol2 transposon transgenesis constructs. Dev Dyn. 2007;236(11):3088–99. pmid:17937395
- 25. Villefranc JA, Amigo J, Lawson ND. Gateway compatible vectors for analysis of gene function in the zebrafish. Dev Dyn. 2007;236(11):3077–87. pmid:17948311
- 26. Bedell VM, Meng QC, Pack MA, Eckenhoff RG. A vertebrate model to reveal neural substrates underlying the transitions between conscious and unconscious states. Sci Rep. 2020;10(1):15789. pmid:32978423
- 27. Patel BB, Clark KL, Kozik EM, Dash L, Kuhlman JA, Sakaguchi DS. Isolation and culture of primary embryonic zebrafish neural tissue. J Neurosci Methods. 2019;328:108419. pmid:31472190
- 28. Kinikoglu B, Kong Y, Liao EC. Characterization of cultured multipotent zebrafish neural crest cells. Exp Biol Med (Maywood). 2014;239(2):159–68. pmid:24326414
- 29. Dutta M, Gilbert SP, Onuchic JN, Jana B. Mechanistic basis of propofol-induced disruption of kinesin processivity. Proc Natl Acad Sci U S A. 2021;118(5):e2023659118. pmid:33495322
- 30. Quintana A, Morgan PG, Kruse SE, Palmiter RD, Sedensky MM. Altered anesthetic sensitivity of mice lacking Ndufs4, a subunit of mitochondrial complex I. PLoS One. 2012;7(8):e42904. pmid:22912761
- 31. Nakajima K, Yin X, Takei Y, Seog D-H, Homma N, Hirokawa N. Molecular motor KIF5A is essential for GABA(A) receptor transport, and KIF5A deletion causes epilepsy. Neuron. 2012;76(5):945–61. pmid:23217743
- 32. Girdhar K, Gruebele M, Chemla YR. The Behavioral Space of Zebrafish Locomotion and Its Neural Network Analog. PLoS One. 2015;10(7):e0128668. pmid:26132396
- 33. Burgess HA, Granato M. Sensorimotor gating in larval zebrafish. J Neurosci. 2007;27(18):4984–94. pmid:17475807
- 34. Eaton RC, Lee RK, Foreman MB. The Mauthner cell and other identified neurons of the brainstem escape network of fish. Prog Neurobiol. 2001;63(4):467–85. pmid:11163687
- 35. Marquart GD, Tabor KM, Bergeron SA, Briggman KL, Burgess HA. Prepontine non-giant neurons drive flexible escape behavior in zebrafish. PLoS Biol. 2019;17(10):e3000480. pmid:31613896
- 36. Allan VJ. Cytoplasmic dynein. Biochem Soc Trans. 2011;39(5):1169–78. pmid:21936784
- 37. Roberts AJ, Kon T, Knight PJ, Sutoh K, Burgess SA. Functions and mechanics of dynein motor proteins. Nat Rev Mol Cell Biol. 2013;14(11):713–26. pmid:24064538