Rm Neves

Modeling the effect of locus coeruleus firing on cortical state dynamics and single-trial sensory processing

Significance Understanding what makes a cortical neuron fire is fundamental to understand how neural circuits operate. We used simultaneous recordings from the locus coeruleus (the neuromodulatory nucleus releasing norepinephrine) and from the somatosensory cortex to formulate a mathematical model explaining how cortical responses originate from the interplay of the sensory drive that cortical neurons receive, the spontaneous dynamics of cortex, and the effect of neuromodulation. Our work provides a hypotheses about how the temporal structure of locus coeruleus burst firing regulates the amplitude and timing of changes in cortical excitability and may selectively amplify responses to salient sensory stimuli. It also suggests that downstream circuits may better decode the activity of a cortical sensory network after estimating its neuromodulatory state. Neuronal responses to sensory stimuli are not only driven by feedforward sensory pathways but also depend upon intrinsic factors (collectively known as the network state) that include ongoing spontaneous activity and neuromodulation. To understand how these factors together regulate cortical dynamics, we recorded simultaneously spontaneous and somatosensory-evoked multiunit activity from primary somatosensory cortex and from the locus coeruleus (LC) (the neuromodulatory nucleus releasing norepinephrine) in urethane-anesthetized rats. We found that bursts of ipsilateral-LC firing preceded by few tens of milliseconds increases of cortical excitability, and that the 1- to 10-Hz rhythmicity of LC discharge appeared to increase the power of delta-band (1–4 Hz) cortical synchronization. To investigate quantitatively how LC firing might causally influence spontaneous and stimulus-driven cortical dynamics, we then constructed and fitted to these data a model describing the dynamical interaction of stimulus drive, ongoing synchronized cortical activity, and noradrenergic neuromodulation. The model proposes a coupling between LC and cortex that can amplify delta-range cortical fluctuations, and shows how suitably timed phasic LC bursts can lead to enhanced cortical responses to weaker stimuli and increased temporal precision of cortical stimulus-evoked responses. Thus, the temporal structure of noradrenergic modulation may selectively and dynamically enhance or attenuate cortical responses to stimuli. Finally, using the model prediction of single-trial cortical stimulus-evoked responses to discount single-trial state-dependent variability increased by ∼70% the sensory information extracted from cortical responses. This suggests that downstream circuits may extract information more effectively after estimating the state of the circuit transmitting the sensory message.

Unilateral electrical stimulation of rat locus coeruleus elicits bilateral response of norepinephrine neurons and sustained activation of medial prefrontal cortex.

The brain stem nucleus locus coeruleus (LC) is thought to modulate cortical excitability by norepinephrine (NE) release in LC forebrain targets. The effects of LC burst discharge, typically evoked by a strong excitatory input, on cortical ongoing activity are poorly understood. To address this question, we combined direct electrical stimulation of LC (LC-DES) with extracellular recording in LC and medial prefrontal cortex (mPFC), an important cortical target of LC. LC-DES consisting of single pulses (0.1-0.5 ms, 0.01-0.05 mA) or pulse trains (20-50 Hz, 50-200 ms) evoked short-latency excitatory and inhibitory LC responses bilaterally as well as a delayed rebound excitation occurring ∼100 ms after stimulation offset. The pulse trains, but not single pulses, reliably elicited mPFC activity change, which was proportional to the stimulation strength. The firing rate of ∼50% of mPFC units was significantly modulated by the strongest LC-DES. Responses of mPFC putative pyramidal neurons included fast (∼100 ms), transient (∼100-200 ms) inhibition (10% of units) or excitation (13%) and delayed (∼500 ms), sustained (∼1 s) excitation (26%). The sustained spiking resembled NE-dependent mPFC activity during the delay period of working memory tasks. Concurrently, the low-frequency (0.1-8 Hz) power of the local field potential (LFP) decreased and high-frequency (>20 Hz) power increased. Overall, the DES-induced LC firing pattern resembled the naturalistic biphasic response of LC-NE neurons to alerting stimuli and was associated with a shift in cortical state that may optimize processing of behaviorally relevant events.

Tracing of noradrenergic projections using manganese-enhanced MRI

We examined the applicability of manganese-enhanced MRI (MEMRI) to the in vivo tracing of diffuse neuromodulatory projections by means of simultaneous iontophoretic injections of an extremely low, non-toxic concentration of MnCl(2) (10mM) and fluorescent dextran in the locus coeruleus (LC) in the rat. We validated the use of the iontophoretic injection by reproducing previously reported results from pressure injections of MnCl(2) in primary somatosensory cortex. Twenty fourhours after injection in LC, Mn(2+) labeling was detected in major cortical and subcortical targets of LC projections including predominantly ipsilateral primary motor and somatosensory cortices, hippocampus and amygdala. Although the injections were in most cases centered in the core of LC, the pattern of Mn(2+) labeling greatly varied across rats. In addition, despite a certain degree of overlap of the labeling obtained with both MEMRI and classical tracing, MEMRI tracing consistently failed to reliably label not only several minor but also major targets of LC, notably the thalamus. The lack of Mn(2+) labeling in thalamus possibly reflected a weaker functional connectivity within coeruleothalamic projections that could not be predicted by anatomical tracing. Inversely, a number of brain regions, particularly contralateral motor cortex, that were not or only sparsely labeled with fluorescent dextran were strongly labeled by Mn(2+). This discrepancy could be partly due to both the activity-dependent and transsynaptic nature of Mn(2+) transport. The overall labeling produced using MEMRI with iontophoretic injections in LC indicates that the Mn(2+) imaging of highly diffuse projections is in principle feasible. However, the labeling pattern of each individual case needs to be carefully interpreted particularly before submitting data for group analysis or in the case of longitudinal examination of discrete changes in functional connectivity under various physiological or behavioral conditions.

Locus coeruleus phasic discharge is essential for stimulus-induced gamma oscillations in the prefrontal cortex

The locus coeruleus (LC) noradrenergic (NE) neuromodulatory system is critically involved in regulation of neural excitability via its diffuse ascending projections. Tonic NE release in the forebrain is essential for maintenance of vigilant states and increases the signal-to-noise ratio of cortical sensory responses. The impact of phasic NE release on cortical activity and sensory processing is less explored. We previously reported that LC microstimulation caused a transient desynchronization of population activity in the medial prefrontal cortex (mPFC), similar to noxious somatosensory stimuli. The LC receives nociceptive information from the medulla and therefore may mediate sensory signaling to its forebrain targets. Here we performed extracellular recordings in LC and mPFC while presenting noxious stimuli in urethane-anesthetized rats. A brief train of foot shocks produced a robust phasic response in the LC and a transient change in the mPFC power spectrum, with the strongest modulation in the gamma (30-90 Hz) range. The LC phasic response preceded prefrontal gamma power increase, and cortical modulation was proportional to the LC excitation. We also quantitatively characterized distinct cortical states and showed that sensory responses in both LC and mPFC depend on the ongoing cortical state. Finally, cessation of the LC firing by bilateral local iontophoretic injection of clonidine, an α2-adrenoreceptor agonist, completely eliminated sensory responses in the mPFC without shifting cortex to a less excitable state. Together, our results suggest that the LC phasic response induces gamma power increase in the PFC and is essential for mediating sensory information along an ascending noxious pathway. NEW & NOTEWORTHY Our study shows linear relationships between locus coeruleus phasic excitation and the amplitude of gamma oscillations in the prefrontal cortex. Results suggest that the locus coeruleus phasic response is essential for mediating sensory information along an ascending noxious pathway.

Monitoring large populations of locus coeruleus neurons reveals the non-global nature of the norepinephrine neuromodulatory system

The non-specific neuromodulation of the forebrain by the noradrenergic locus coeruleus (LC) is a foundation of wide-ranging theories of cognitive and systems neuroscience. The non-specificity is assumed because of the diffuse projections of the nucleus as well as the synchronous spiking of its neurons. Synchrony, however, has never been assessed in a large population of LC cells, i.e. single units, nor has it been systematically related to specificity of their projection targets. Here, we recorded up to 52 single units simultaneously (3164 unit pairs) in the rat LC, and characterized forebrain projection patterns using antidromic stimulation. Two novel unit types were characterized; they differed by waveform shape, firing rate, and propensity for synchronization. Cross-correlation analysis revealed a surprisingly poor correlation between unit spiking; only 13% of unit pairs had response profiles reflecting synchronization due to common synaptic input or gap junctions. While LC unit spikes were phase locked to cortical slow oscillations (< 2 Hz), they did so intermittently, yielding poor population synchronization. A novel infra-slow (0.01-1 Hz) spiking fluctuation was observed in LC units, yet this too was asynchronous across unit pairs. A highly synchronized minority had a stronger tendency for targeted forebrain neuromodulation. Our findings demonstrate that the LC may convey a more complex and differentiated neuromodulatory signal than is widely assumed.

A dynamical systems model of the effect of Locus Coeruleus firing on single trial cortical state dynamics

Activity of sensory areas continuously varies reflecting both changes in external sensory stimuli and in internal states of the organism that do not necessarily always have a precise relationship to sensory inputs. An important aim in systems neuroscience is to develop quantitative models that may explain how state-dependent representations are generated and how they can be best interpreted. Statedependent modulations of cortical activity are in part mediated by changes in activity of various subcortical neuromodulatory nuclei. A prominent example is the Locus Coeruleus (LC), which can modulate both ongoing changes of cortical states and their responses to sensory inputs. However, a quantitative model of how the temporal fluctuations of LC firing affect ongoing and stimulus-driven primary cortical dynamics is currently missing. Here we investigated how LC modulates the ongoing cortical states and the sensory information carried by cortical firing using a combination of neurophysiological experiments and data-driven dynamic-system models of cortical state changes. We performed simultaneous recordings of neural activity in primary somatosensory cortex and both ipsilateral and contralateral LC in urethane anaesthetized rats during spontaneous activity and during electrical stimulation of the contralateral hind paw. On the basis of this data, we have constructed a novel data-driven dynamical system model of cortical states dynamics. This model extends and generalizes recent simple effective models [1] by including the effect of firing of LC noradrenergic neurons. We first fitted dynamic systems models of cortical states that either did or did not contain LC-cortical interactions. We found that models omitting the LC noradrenergic input to cortex tend to describe the dynamics of spontaneous activity of S1 cortex reasonably well. However, including ipsilateral (and to a lesser extent contralateral) LC activity as input to the models make the prediction of cortical states and of single trials responses much better. We then investigated which aspects of the LC dynamics help the model to increase cortical state predictability. We found that ipsilateral LC firing activity at low frequencies (< 10 Hz) correlated positively with slow (1-6 Hz) fluctuations of cortical power. The insertion of ipsilateral LC input to the model captured this dynamics by creating additional low frequency (1-6 Hz) state variation of model activity that correlated, both in power and phase, to those observed in real cortical activity. We finally investigated how these dynamical systems models can be used to predict single trial sensory evoked responses and to understand how this dynamics shapes the information representation of sensory stimuli. We derived an explicit mathematical rule predicting the trial-by-trial variability of cortical responses to stimuli arising from LC-modulated cortical state dynamics, and found that subtracting this variability from single trial cortical responses approximately doubles the amount of mutual information about the somatosensory stimuli that could be extracted from cortical responses.