A Cognitive-Motor Dissociation of NMDAR Blockade in Behaving Non-Human Primates

doi:10.21203/rs.3.rs-3124207/v1

Download PDF

Article

A Cognitive-Motor Dissociation of NMDAR Blockade in Behaving Non-Human Primates

https://doi.org/10.21203/rs.3.rs-3124207/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Ketamine is widely used to probe cognitive functions relying on the properties of NMDAR synaptic transmission. Numerous works have already proved that cognitive performance and adjustments in the decision or perceptual domains are affected after ketamine injection in the general circulation. Here, we take advantage of that in the brain stem; horizontal saccade deceleration is controlled by glycine NMDA-gated current. In contrast, GABA current controls vertical deceleration to demonstrate that despite general circulation level manipulation of NMDAR synaptic transmission, the skewness of the saccade appeared to be in the motor brainstem generator circuit differentially maintained. The deacceleration of the saccade elicited toward a horizontal target was substantially decreased, while the deacceleration of a vertical saccade remained largely unaffected. These results provide a functional marker for estimating NMDA-gated specificity acting in the pre-frontal cortex while maintaining its specificity among the GABA circuit of drugs in general circulation.

Biological sciences/Drug discovery

Biological sciences/Neuroscience

Biological sciences/Psychology

At a subanesthetic dose (IM: 0.5-2mg/kg), Ketamine, an N-methyl-D-aspartate receptor antagonist, interferes with multiple aspects of cognition in monkeys^1–3. Ketamine has been shown to disturb decisions when rules have not been learned³, reduces task selectivity in macaque dorsolateral prefrontal neurons, and impairs the performance of randomly interleaved prosaccades and antisaccades or task switching ^{4,5 6}. At the oculomotor level, smaller doses (0.1-0.5mg/kg) have been shown to benefit monkeys on visual search tasks and saccade-delayed tasks ^7,8. Human subjects show a paradoxically increased performance in target tracking, which unexpectedly changes direction (Allen et al., 2009), but the gain of smooth pursuit is strongly altered⁹. Humans also failed to exploit environmental regularities in saccade tasks¹⁰. The dynamics of the saccades have been reported to be not altered (the amplitude/maximum velocity relationship) by ketamine at the dose of 1 mg/kg.)^11,12. These results were surprising because chemical lesions of the brain stem region containing glycinergic omnipause neurons (OPNs) cause saccade slowing with no change in latency. Simulation studies were performed with a conductance-based model of premotor excitatory burst neurons (EBNs) that incorporated multiple membrane channels, including the T-type calcium channel, to explore the mechanisms responsible for this deficit¹³. The peak speed of a typical saccade was determined by the T- and NMDA currents in EBNs after the OPNs shut off. After OPN lesions, the model made slow saccades because the EBN activity was lower than expected due to a reduced T-current (caused by the loss of hyperpolarization) and a reduced NMDA current (caused by a reduced glycine concentration around the receptors).

In the present study, we aimed to characterize the dissociative effect of ketamine intake between a cognitive task and saccades kinematics. To achieve these goals, we analyzed a previously acquired ⁷extensive set of data, which is a rare possibility given the short time window of ketamine effects. Thus, we investigated the possibility that the observed effects of ketamine on saccade dynamics could rely solely on the modulation of premotor structures activity, the excitatory burst neurons. We finally tested our hypothesis using a model to generate saccades ¹³.

This paper compares saccade kinematics with and without intramuscular ketamine injection while monkeys executed a visual search task. Data were acquired by Shen et al. (2010). Methods from Shen and Paré (2006) are summarized here. All animal care and experimental protocols were approved by the Queen's University Animal Care Committee and were in accordance with the Canadian Council on Animal Care guidelines. the study is reported in accordance with ARRIVE guidelines.

Experimental setup:

Stimuli were built on a personal computer running the Psychophysics Toolbox on MATLAB (The MathWorks, Natick, MA; Brainard, 1997). They were displayed on a 37inches monitor (Mitsubishi XC- 3730C Megaview Pro 37) located 57 cm from the animal. Monkeys sat in primate chairs with their heads restrained. Positions of the eyes were recorded (500Hz) with an infrared camera (EyelinkII, SR Research Ltd., Mississauga, Ontario, Canada). Paradigms were displayed by a real-time QNX-architecture REX system⁷.

Paradigm:

Each trial began with the appearance of a central, white, 0.5°-wide fixation spot. Monkeys were required to fixate the target for 1s. Then, eight spots were displayed equally spaced along a 10 deg radius circle. Seven had a low luminance of 10.6 cd/m2, while the eighth one had a higher luminance varying randomly from trial to trial between 10.9 and 76.6 cd/m2. Within 500ms, monkeys had to look at the most luminous dot (2x2 deg validation window). They had to maintain fixation for 200ms to be rewarded. A trial was labeled as correct if the first saccade was initiated toward the target and if the monkey stayed in the validation window for at least 200ms. If the monkey reached the right target despite an erroneous initial saccade, the trial was labeled incorrect (monkeys were rewarded with a smaller amount of water). If the target was not reached within 2000ms following its display, monkeys were not rewarded, and the next inter-trial was lengthened. Each session consisted of two blocks of 500–800 trials, lasting approximately 15–20 minutes. Injections were done between the first and the second block of each session. Thus, block one assessed the baseline performance of the animal, and block 2 assessed the specific role of the injection.

Drug:

Three doses of ketamine (0.25, 0.5, 0.75 mg/kg) and a saline solution (0.9% NaCl) as a control were used in this study. All were injected intramuscularly. Although ketamine is widely known and used as a dissociative anesthetic, numerous studies reported that doses < 1.0 mg/kg do not cause debilitating effects on behavior^14,15. Sub-anesthetic doses of ketamine (< 1.5mg/kg) could induce a tolerance if injections are frequently repeated¹⁶. Thus, monkeys were injected once a week, except for the lowest dose, which was sometimes tested twice weekly. Three injections per dose were performed in a randomized order (12 injections per monkey).

Behavior analysis:

Eye movements from three female rhesus monkeys (Macaca mulatta, G, H, and F) were recorded⁷. H was excluded from the present study because she was tested with one medium dose of ketamine. Queen’s University Animal Care Committee approved animal care and experimental protocols by the Canadian Council of Animal Care guidelines. Data were analyzed offline using a custom-made Matlab script. Saccades were detected using a velocity threshold criterion (30°/s). Eye velocity was computed by differentiation and low-pass filtering of digitized eye position (Usui and Amidror, 1982). The time interval of the ketamine effect was chosen from the second minute to the twentieth. Indeed, latency was significantly affected approximately 2 minutes after the injection (see (Shen et al., 2010), Figs. 2A-B). The effect lasted approximately 20 minutes. Our post-doc analysis included the first saccade of correct trials occurring within this time window (18 minutes wide) for the treatment block of each session. Eight thousand two hundred sixty-four saccades were included in this study, which corresponds to 363 ± 115 trials per orientation, per monkey, and dose. We pooled the data corresponding to the left-right horizontal orientations, the up-down vertical ones, and the oblique ones. We can assume the brainstem saccade generator is somewhat equivalent within each orientation pair. Besides, the literature does not report systematic differences between the kinematics of leftward versus rightward and upward versus downward saccadic eye movements^17,18. We verified this assumption by comparing the amplitudes of leftward versus rightward and upward versus downward saccades. None of them is significant at a corrected α-threshold of 0.05.

As shown in Fig. 1, we computed the latency, peak and average velocity, amplitude, duration, and skewness for each saccade. To evaluate the skewness, we computed the ratio between the duration of acceleration (between saccade onset and time of peak velocity) and deceleration (between the time of peak velocity and saccade offset). Thereby, a skewness of 1 indicates a symmetrical velocity profile. A skewness smaller than 1 corresponds to a more extended deceleration phase and asymmetrical velocity profiles. Finally, skewness more significant than one corresponds to a more extended acceleration phase.

Model and Simulations:

Simulations were performed with SIMULINK on a portable computer. We used a fixed-step Runge-Kutta-4 simulation solver with a step size of 1µs. We made a simple adjustment of NMDA currents to model the effects of a decreasing horizontal skewness with an increasing dose of ketamine. Their outputs (INMDA) were multiplied by a fixed gain, hypothesized to be the degree of blockade of the NMDA channels. Four linearly decreasing gains were applied to the output of NMDA channels: 1.0, 0.66, 0.33, and 0.0. We simulated three sizes of saccades: 10°, 15°, and 20°.

As presented in Fig. 2, we replicated the model of Miura¹³, which seemed to be of a reasonable level of detail¹⁹. This is a hybrid control model with feedback on the current eye position. It incorporates two one-compartment conductance-based neurons integrated into a simple and standard lumped model of saccade generation. Therefore, the focus falls on the contribution of the membrane potential of this neuron to the eye movement, which is the output of the whole system. It is worth noting that this model generates only horizontal eye movements.

Apart from the comparator (on the left of the previous diagram) and the latch circuit (lying inside the OPN unit), which have yet to find an anatomical substrate, every block of this diagram has a functional basis. Although, a significant part of the BSG needs to be more concise or present in the model. For example, the neural integrator for eye position is perfect, contradicting experimental data²⁰. The integrated eye position is the same signal as the muscle command and the feedback signal. And finally, the viscoelastic properties of the muscles are not incorporated.

Input and feedback to the EBN: The model input is the desired eye displacement (ed). That command is fed to a comparator, which subtracts the efferent copy eye position to this input sending to the downstream structures a dynamic estimate of instantaneous motor error (êm). The model is symmetric downstream to the comparator. The two branches contain the same elements, differing only by the input signal they receive: one receives a positive drive if the instantaneous motor error is positive, i.e., the eye has not reached the target yet; the second receives a positive drive if the instantaneous motor error is negative, which appeared to be the case when the eye overshoots the target. Negative drives are rectified to zero, and positive drives are fed to the EBN unit through a low-pass filter (5ms time constant), reflecting the “sluggish” development of the activity of LLBN (providing the actual drive to the EBN)²¹.

The EBN unit is simulated as a conductance-based membrane whose output is a transfer function generating spike trains. The following classical equation describes the membrane voltage: Apart from the notable exception of the Leak current, whose conductance stays constant with voltage variations, other currents are defined by dynamical systems. Dynamical equations of the currents can be found in the original paper¹³. The EBN output transforms the membrane potential in a signal reflecting its state: Thus, whenever the EBN potential is over − 10mV, the EBN output is set to 1.

The OPN unit

This system includes one OPN unit and projects to both EBNs. The state of the OPN unit is defined by the sum of a constant bias (= 1), a brief inhibitory trigger (20 ms inhibitory pulse = − 2) starting when the drive input reaches the EBNs, and an inhibitory latch signal. The magnitude of the latch signal is defined as the sum of outputs of the two EBNs multiplied by 100 and fed through a low-pass filter (time constant 50 ms). The state was rectified as either positive or null, and then this signal was fed to the glycine channels of both EBNs. This output of the two EBNs is summed and then multiplied by a factor K, reflecting the exploration of the eye caused by one EBN spike. It has been set to 4,5. Eventually, this signal is fed to a mathematical integrator, issuing an eye displacement signal. This last signal is used as feedback and input to the extraocular muscles.

All data were subsequently sub-sampled to 1000Hz. Eye position data issued by the model were treated with the same differentiation methods and filter algorithms as those used on our second data set (see Methods). EBN currents and potentials were filtered after clipping off the variations related to the action potentials, as described in the original paper, with a digital three-pole Butterworth low-pass filter (cut-off frequency of 30Hz).

This section briefly describes the results obtained in Shen’s covert visual search task⁷. We did not reanalyze the performances.

From previous analysis a question has arisen: is this effect different from orientation to orientation? Is the effect of ketamine spatially selective? Fig. 2 supports a negative answer to this issue; ketamine seemingly has a similar effect on reaction times, regardless of the target’s position on the screen. As previously stated, we questioned ourselves about the outcome of the ketamine on the oculomotor saccadic system for at least three reasons: they have never been fully described; these experiments are of considerable importance considering the highly tight time-window effect of ketamine (about fifteen to twenty minutes); given the clustered organization of the BSG, ketamine is an attractive probe to its function.

As shown in Fig. 4 for horizontal saccades, the ketamine decreases the saccade skewness, in both tested monkeys, in a dose-dependent fashion. All statistical tests were two-tailed t-tests. The difference in skewness is significant (p < 0.001, Bonferroni-corrected) for each pair of ketamine doses (statistics tables not shown). In oblique saccades, the ketamine also decreases the saccade skewness significantly in both tested monkeys in a dose-dependent fashion, but with a lower influence (p < 0.001 Bonferroni-corrected, for each pair of ketamine doses, except between the lower doses significant at p < 0.01; statistics tables not shown). However, the ketamine appeared not to influence vertical saccades regarding skewness (one significant difference between the two middle doses, p < 0.01) in both monkeys. These results contrast sharply with the cognitive effect of ketamine, as shown by latency reciprobit plots: we are facing a cognitive-motor dissociation of ketamine effects. Eventually, we report a well-known effect of ketamine on the amplitude of the saccades (decreased with the dose, Supplementary Fig. 1), which is accompanied by a post-saccadic drift. Both have yet to be further studied.

Finally, a widespread way to compare abnormal saccades metrics is to plot the primary sequence²². Because a parametric relationship holds between saccade amplitude, peak velocity, and saccade duration, any difference in the main sequence plot reveals an underlying anomaly in the BSG. Figure 3 shows an actual plot of the primary sequence of these data. For each amplitude bin (0.5 degrees wide), peak velocity and mean velocity of saccades are significantly decreased in the high-dose condition compared to the saline control test. Saccade duration is significantly increased after ketamine injection. The previous results of the modulation of saccade skewness by ketamine led us to conclude a fine modulation of premotor structure-activity. Two main results are to be explained. First, how the perturbation of the conductance of one particular channel could modulate the output of an entire system? Second, how could we account for the orientation selectivity of this effect, which is inexistent for vertical saccades but highly marked for horizontal saccades?

As displayed in Supplementary Fig. 1, blocking the current flowing through the NMDAR has consequences for smoothing and curving the trajectory of the saccade, regardless of its programmed amplitude. The more the NMDAR is blocked, the curvier the saccade is. The adjunction of an NMDA channel blocker, like ketamine, has relatively little impact on both the saccade velocity and the EBN spikes train. As presented in Supplementary Fig. 2, in this model, for 10° saccades, the contribution of NMDA-gated currents to the total EBN current inflow is probably too small for an alteration of its function to have severe consequences for the overall dynamics. For 15°, though, blocking the NMDAR exhibits some noticeable outcomes on EBN firing and, thus, saccade velocity. The increased NMDA block causes the spikes train to somewhat “stretch”: the number of spikes emitted by the EBN unit remains, but their temporal display expands. The firing rate decreases, inverse of the Inter-Spike Interval (ISI). The original paper showed that the EBN firing rate was related to the velocity. Here, we can note a decrease in the peak- velocity of the simulated saccade due to ISIs lengthening. The last observation is the increase in duration caused by an overall more extended activity of the EBN. The same results are observed for larger amplitudes when blocking the NMDAR: ISI lengthening with endless spikes, decrease in peak velocity, and increase in duration.

The skewness of these saccades is modified dose-related, as expected from the behavioral experiments. But, the model sheds new light on the influence of NMDAR for large amplitude saccades: its participation in the total current increases non-linearly with the saccade size¹³. The decrease in skewness is also amplitude-dependent. An additional plot of skewness modulation seemed unnecessary since it can directly be read on the velocity plots.

We have confirmed that saccade latencies in a visual search task are dose-related. This decrease is independent of the direction in which the correct saccade is made. Importantly, saccade metrics are affected in significantly different ways regarding their direction. Indeed, the skewness, an index characterizing the shape of the velocity profile of saccades, is mainly perturbed in horizontal saccades and unchanged for vertical saccades. Furthermore, saccades made after the injection of ketamine could be defined by a slightly different main sequence: their peak- and mean velocities are decreased, and their duration is increased. The first result of the cognitive effect of ketamine implies that the NMDAR has an essential role in decision-making. We have already reviewed the pieces of evidence for single-neuron properties of perceptual accumulators. They probably reflect the activity of an immensely more extensive neural network implicated in this kind of sensory decision-making. Xiao-Jing Wang was among the first to propose the idea that the accumulation properties of a network could be seen as an emergent property of the particularity of NMDAR synaptic transmission^23–25. Indeed, former neural decision-making models were frequently constructed as bistable state networks. The bistability of these networks could accurately capture the temporal transition between two cognitive states but fail to capture the transition between the previous state to the subsequent one²⁶. This transition is the essence of accumulator models, which display a gradually increasing activity. Deciding on a noisy, spiking, conductance-based network relies on the dynamics of NMDAR²⁵: they open and close with a slow dynamic, enabling a gradual gating of neuronal excitability. Increasing a neuron’s excitability makes it prone to respond with a higher discharge rate, as it permits maintaining activity by making excitatory feedback connections inside the network. Our present work does not demonstrate that varying the gain of NMDAR current decreases the firing rate of simulated neurons in a decision-making network; it provides substantial clues, however.

Besides, this modulation of the NMDAR gain in a realistic network has already been published ²⁷ and show similar results in a sensory discrimination task. Moreover, the dissociation observed between the unidirectionality of cognitive ketamine effects and the direction-dependent effect on saccade metrics led us to the hypothesis that this difference should emerge in brain areas where saccades are not encoded as a unified desired displacement vector (like FEF or SC) but as separate horizontal and vertical components. We provided a hypothesis for explaining this dissociation. In the brainstem, saccade components are encoded in different anatomical structures. These structures receive different inhibitory projections from other premotor brainstem areas; the horizontal bursters nucleus (PPRF) receives more glycinergic inputs than the vertical one (riMLF). Yet, the glycine is a co-agonist of the NMDAR and could spill over inhibitory synapses to other ones, substantially modifying the excitability of the neurons bathed in glycine. As a result, because ketamine is a use-dependent antagonist, horizontal EBNs are more affected by NMDAR antagonization than vertical EBNs. The model was replicated to understand how the modulation of an NMDA-gated current in EBN could be sufficient to increase the asymmetry of a saccade. We could replicate most behavioral findings with a specific adaptation of a hybrid (conductance-based EBN element included in a lumped circuit) model¹³. Indeed, blocking NMDA-current resulted in increased saccade duration, accompanied by a decrease of both peak- and mean velocity. These effects depended on the saccade size, which was absent from our analysis because only one amplitude had been tested.

It was not previously thought that premotor regions could be functionally dissociated by injecting a pharmacological substance with such a broad activity spectrum. Thus, a challenge is to inject a drug that preferentially lowers the activity of vertical bursters regions and modifies the asymmetry of vertical saccades more than horizontal ones. This drug cocktail is thought to be a GABA antagonist. Another preliminary study would be the replication of the present findings. As proposed in a recent and elegant study that deductions can be made about electrophysiological firing properties by histochemical examination of functional groups within the brainstem saccadic circuitry. Our functional finding is another building block supporting the concept that channelopathies in saccadic disorders may be important to consider. Future histological and neuro-compatible modelling studies in humans might help to refine and confirm this approach for saccadic disorders²⁸. Moreover, our investigation of motor consequences in a broader range of angles and eccentricities would help to characterize the specific effects of NMDAR blockade on saccade metrics. Eventually, ketamine is an excellent probe to test the reliance of decision-making on NMDAR, as supposed by modeling experiments. Sufficient behavioral and computational data arose to propose the idea of an electrophysiological test of this hypothesis.

This worked based on post hoc analysis of a large dataset and the replication of a model for saccade generation to investigate cognitive and motor consequences of NMDAR blocker intake. In a visual search task, general injection of the drug produced effects on latencies following the reduction in the rate of rise of a perceptual accumulator for decision-making. These cognitive effects were independent of the target's position on the display. However, the skewness of the saccades made after ketamine injection appeared to be differentially modulated: the horizontal ones were significantly decreased, while the vertical ones remained unchanged. A possible interpretation of these findings is based on local glycine levels and theories of glycine gating activity. Finally, all saccades’ parameters modified by injection of ketamine found an explanation at the level of the EBN membrane. We opened the questions about the possibility of finding the same dissociation for vertical saccades with other pharmacological agents and whether build-up neurons’ activity crucially relies on the NMDA receptor.

Our results provided a physiological basis for dissociating ketamine effects on saccade skewness, even though it has not been injected locally. The premotor burster neuron activity ²⁹ correlates exceptionally well with the instantaneous saccadic velocity and the dynamic motor error. It is believed that the EBNs provide the saccadic pulse and velocity drive to the ipsilateral ocular motoneuron. Thus, modifying its discharge pattern could lead to modifying the saccade kinematics. Another remarkable fact about the bursters neurons lies in the difference in their neurotransmitters: EBNs use glutamate; hIBNs use glycine; vIBNs use GABA. This is a fundamental functional difference between the horizontal and the vertical/torsional saccade generator systems. Optican was among the first to propose a functional role for glycinergic transmission in the BSG^13,30. An extraordinary property of the NMDAR is to be co-activated by two different substrates: Glutamate and Glycine.

Furthermore, it has been proved that the glycine released in a synapse could have –enhancing- effects on the NMDAR of other synapses³¹. Thus, glycine could modulate the excitability of a pool of neurons: the more glycinergic inputs a neuron receives, the more excitable it would be. Differences in glycine concentration “around” different neurons would mean a difference in NMDAR dynamics while receiving excitatory inputs. This is our proposal to explain the dissociation of skewness modulation by ketamine. During fixation, all bursters (vertical and horizontal) neurons are tonically inhibited by glycinergic projections coming from the OPN. During saccades, the EBNs (vertical and horizontal) activate the ipsilateral OMN, while their respective ipsilateral IBNs phasically inhibit the contralateral OMN. Finally, at the end of a saccade, the contralateral IBNs choke off the ipsilateral EBN. Yet, this last burst of activity is conveyed by different neurotransmitters according to the direction of the saccade. Since the time constant of the glycine spillover effect is about a minute³¹, which is mainly superior to the time constant of EBN channel dynamics, we hypothesize that horizontal EBNs are constantly bathed in a higher concentration of glycine than do vertical IBNs. Although it is a rough hypothesis since we cannot dynamically probe glycine's extracellular concentration around EBN, it is biophysically plausible³¹.

Author contributions.

M.P. collected the original data. P.P and, P.D. design model and analysis of acquired data. P.P and P.D. modelized the data. P.P. M.P. and D.P. supervised this project. Abstract, core text as well as figures were done by P.P. P.D. and M.P.

Data availability. The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.

Competing interests. P.P. and P.D. are co-inventors of several patents in eye movement tracking and co-founders of P3Lab company which commercializes clinical eye-tracking systems. M.P. has nothing to disclose.

Taffe, M. A., Davis, S. A., Gutierrez, T. & Gold, L. H. Ketamine impairs multiple cognitive domains in rhesus monkeys. Drug Alcohol Depend. 68, 175–187 (2002).
Weed, M. R. et al. Negative Allosteric Modulators Selective for The NR2B Subtype of The NMDA Receptor Impair Cognition in Multiple Domains. Neuropsychopharmacology 41, 568–577 (2016).
Brunamonti, E. et al. The NMDAr antagonist ketamine interferes with manipulation of information for transitive inference reasoning in non-human primates. J. Psychopharmacol. (Oxf.) 28, 881–887 (2014).
Skoblenick, K. & Everling, S. N -methyl- d -aspartate Receptor Antagonist Ketamine Impairs Action-monitoring Activity in the Prefrontal Cortex. J. Cogn. Neurosci. 26, 577–592 (2014).
Skoblenick, K. & Everling, S. NMDA Antagonist Ketamine Reduces Task Selectivity in Macaque Dorsolateral Prefrontal Neurons and Impairs Performance of Randomly Interleaved Prosaccades and Antisaccades. J. Neurosci. 32, 12018–12027 (2012).
Stoet, G. & Snyder, L. H. Effects of the NMDA Antagonist Ketamine on Task-Switching Performance: Evidence for Specific Impairments of Executive Control. Neuropsychopharmacology 31, 1675–1681 (2006).
Shen, K., Kalwarowsky, S., Clarence, W., Brunamonti, E. & Pare, M. Beneficial Effects of the NMDA Antagonist Ketamine on Decision Processes in Visual Search. J. Neurosci. 30, 9947–9953 (2010).
Ilhame, A., Nicolas, W., Pierre, P. & Marcus, M. A subanesthetic dose of ketamine in the Rhesus monkey reduces the occurrence of anticipatory saccades. Psychopharmacology (Berl.) 232, 3563–3572 (2015).
Thakkar, K. N., Diwadkar, V. A. & Rolfs, M. Oculomotor Prediction: A Window into the Psychotic Mind. Trends Cogn. Sci. 21, 344–356 (2017).
Avila, M. T., Hong, L. E., Moates, A., Turano, K. A. & Thaker, G. K. Role of Anticipation in Schizophrenia-Related Pursuit Initiation Deficits. J. Neurophysiol. 95, 593–601 (2006).
Godaux, E., Cheron, G. & Mettens, P. Ketamine induces failure of the oculomotor neural integrator in the cat. Neurosci. Lett. 116, 162–167 (1990).
Krystal, J. H. Subanesthetic Effects of the Noncompetitive NMDA Antagonist, Ketamine, in Humans: Psychotomimetic, Perceptual, Cognitive, and Neuroendocrine Responses. Arch. Gen. Psychiatry 51, 199 (1994).
Miura, K. & Optican, L. M. Membrane channel properties of premotor excitatory burst neurons may underlie saccade slowing after lesions of omnipause neurons. J. Comput. Neurosci. 20, 25–41 (2006).
Condy, C., Wattiez, N., Rivaud-Péchoux, S. & Gaymard, B. Ketamine-induced distractibility: An oculomotor study in monkeys. Biol. Psychiatry 57, 366–372 (2005).
Shiigi, Y. & Casey, D. E. Behavioral effects of ketamine, an NMDA glutamatergic antagonist, in non-human primates. Psychopharmacology (Berl.) 146, 67–72 (1999).
Pouget, P., Wattiez, N., Rivaud-Péchoux, S. & Gaymard, B. Rapid development of tolerance to sub-anaesthetic dose of ketamine: an oculomotor study in macaque monkeys. Psychopharmacology (Berl.) 209, 313–318 (2010).
Collewijn, H., Erkelens, C. J. & Steinman, R. M. Binocular co-ordination of human vertical saccadic eye movements. J. Physiol. 404, 183–197 (1988).
Collewijn, H., Erkelens, C. J. & Steinman, R. M. Binocular co-ordination of human horizontal saccadic eye movements. J. Physiol. 404, 157–182 (1988).
Gerstner, W., Sprekeler, H. & Deco, G. Theory and Simulation in Neuroscience. Science 338, 60–65 (2012).
Becker, W. & Klein, H.-M. Accuracy of saccadic eye movements and maintenance of eccentric eye positions in the dark. Vision Res. 13, 1021–1034 (1973).
Scudder, C. A., Fuchs, A. F. & Langer, T. P. Characteristics and functional identification of saccadic inhibitory burst neurons in the alert monkey. J. Neurophysiol. 59, 1430–1454 (1988).
Zee, D. S. & Robinson, D. A. A hypothetical explanation of saccadic oscillations. Ann. Neurol. 5, 405–414 (1979).
Wang, M. et al. NMDA Receptors Subserve Persistent Neuronal Firing during Working Memory in Dorsolateral Prefrontal Cortex. Neuron 77, 736–749 (2013).
Wang, X.-J. Probabilistic Decision Making by Slow Reverberation in Cortical Circuits. Neuron 36, 955–968 (2002).
Wang, X. J. Synaptic basis of cortical persistent activity: the importance of NMDA receptors to working memory. J. Neurosci. Off. J. Soc. Neurosci. 19, 9587–9603 (1999).
Goldman, M. S. Robust Persistent Neural Activity in a Model Integrator with Multiple Hysteretic Dendrites per Neuron. Cereb. Cortex 13, 1185–1195 (2003).
Standage, D. & Paré, M. Persistent storage capability impairs decision making in a biophysical network model. Neural Netw. 24, 1062–1073 (2011).
Mayadali, Ü. S., Lienbacher, K., Shaikh, A. G. & Horn, A. K. E. Saccadic premotor burst neurons and histochemical correlates of their firing patterns in rhesus monkey. J. Neurol. Sci. 439, 120328 (2022).
Van Gisbergen, J. A., Robinson, D. A. & Gielen, S. A quantitative analysis of generation of saccadic eye movements by burst neurons. J. Neurophysiol. 45, 417–442 (1981).
Optican, L. M. The role of omnipause neurons: why glycine? in Progress in Brain Research vol. 171 115–121 (Elsevier, 2008).
Ahmadi, S. et al. Facilitation of Spinal NMDA Receptor Currents by Spillover of Synaptically Released Glycine. Science 300, 2094–2097 (2003).

Competing interest reported. P.P. and P.D. are co-inventors of several patents in eye movement tracking and co-founders of P3Lab company which commercializes clinical eye-tracking systems. M.P. has nothing to disclose.

Download PDF

Version 1

posted

You are reading this latest preprint version

A Cognitive-Motor Dissociation of NMDAR Blockade in Behaving Non-Human Primates

Status:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Experimental setup:

Paradigm:

Drug:

Behavior analysis:

Model and Simulations:

Results

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1