Targeting modulation of noradrenalin release in the brain for amelioration of REMS loss-associated effects.

Rapid eye movement sleep (REMS) loss affects most of the physiological processes, and it has been proposed that REMS maintains normal physiological processes. Changes in cultural, social, personal traits and life-style severely affect the amount and pattern of sleep, including REMS, which then manifests symptoms in animals, including humans. The effects may vary from simple fatigue and irritability to severe patho-physiological and behavioral deficits such as cognitive and behavioral dysfunctions. It has been a challenge to identify a molecule(s) that may have a potential for treating REMS loss-associated symptoms, which are very diverse. For decades, the critical role of locus coeruleus neurons in regulating REMS has been known, which has further been supported by the fact that the noradrenalin (NA) level is elevated in the brain after REMS loss. In this review, we have collected evidence from the published literature, including those from this laboratory, and argue that factors that affect REMS and vice versa modulate the level of a common molecule, the NA. Further, NA is known to affect the physiological processes affected by REMS loss. Therefore, we propose that modulation of the level of NA in the brain may be targeted for treating REMS loss-related symptoms. Further, we also argue that among the various ways to affect the release of NA-level, targeting α2 adrenoceptor autoreceptor on the pre-synaptic terminal may be the better option for ameliorating REMS loss-associated symptoms.

INTRODUCTION

Rapid eye movement sleep (REMS) is a unique behavioral phenomenon and is an integral component of sleep. Different states of sleep and wakefulness have been objectively identified by the simultaneous presence or absence of associated classical electrophysiological signals recorded from the brain, the electroencephalogram (EEG), eye movements, the electrooculogram (EOG) and muscle tone and the electromyogram (EMG). REMS is classically and objectively identified and quantified by desynchronized EEG, atonia in the antigravity muscles, rapid eye movements and appearance of pontogeniculo occipital (PGO) waves. Thus, as compared with waking, during REMS the EEG and EOG apparently resemble that associated with waking, while EMG shows an opposite expression. Therefore, this sleep state has also been referred to as “active sleep” or “paradoxical sleep” or “desynchronized sleep.” Further, as this stage is often associated with dreaming, this stage has been termed as dream state of sleep, although it is known that sometimes dream may appear during non-REMS as well. The REMS has been identified in almost all the higher species in evolution recorded so far, including humans.[ To this effect, we reiterate our argument that as one of the primary characteristic features to identify REMS is the recording from the brain the EEG signals, consequently this stage of sleep has been identified in species higher in evolution possessing developed and evolved brain. As a corollary, it is also obvious that because of a lack of fundamental biomolecular markers, it is yet to be confirmed if this stage is present, may be even in a rudimentary form in other lower species.

The quantity of REMS varies among species; it is affected by life-style changes and under various psycho-somato-patho-physio-logical altered conditions.[ Under normal conditions, the quantum of REMS reduces with ageing; however, normally, this stage is never absent through life.[ The essentiality of REMS to maintain normal life processes may be supported by the fact that it is not under voluntary control, it is affected in almost all disorders, its loss is followed by a compensatory rebound increase during the post-deprivation recovery period, its prolonged loss may be fatal and, finally, its basic regulatory mechanism lies at the core of the brainstem as that of controlling other absolutely essential functions for living, e.g. heart rate, respiration, blood pressure and sleep-waking, which, however, may be modulated by other factors.

Involvement of noradrenaline (NA) in REMS-loss associated symptoms

REMS loss is often associated with expressions of a number of specific and nonspecific symptoms, many of which may or may be apparently unrelated. Although it may be difficult to associate REMS loss as an exclusive cause or effect of a disorder, we propose that in the case of REMS disturbance (directly or indirectly as a cause or an effect), the neural mechanism regulating it must be affected. Therefore, targeting a common factor, if any, which is known to be essentially expressed during REMS as well as under conditions of REMS loss, may be targeted to help the sufferer. Altered REMS is one of the common symptoms associated with many disordered states, including hypertension,[ hyperglycemia, hyper-excitability, [ lack of concentration, memory loss,[ Alzheimer’s disease,[ Parkinson’s disease[ and depression.[ The level of NA is altered in most of these altered states (dysfunctions) and, interestingly, NA is elevated after REMS loss as well.[ By and large, NA is at least an important common causative factor inducing many of the REMS deprivation (REMSD)-associated symptoms, including neuronal cytomorphometry,[ apoptosis in brain neurons,[ increased Na–K ATPase activity[ and thermoregulatory changes[ that were prevented by adrenoceptor (AR) antagonists.[ Therefore, we propose that targeting mechanisms(s) that maintain the NA level in the brain could be a possible way to address treatment of REMS loss-associated symptoms. However, before we proceed to justify in support of our proposition, what role NA plays in REMS regulation needs to be evaluated.

Neural regulation of REMS and its relation with brain level of NA

REMS has a well-regulated cyclic appearance; at least in humans, its frequency and duration per episode increases with the depth of sleep.[ Lesion, transection and stimulation studies have shown that neurons located in the brainstem are its primary regulators, while neurons in other brain regions modulate REMS by influencing these primary regulators.[ On the basis of temporal correlation of firing rate of neurons during REMS, neurons have been classified as REM-ON (those are active during REMS) or REM-OFF (those are silent during REMS).[ It is classically known that REM-ON neurons are presumably ACh-ergic and are concentrated largely within the latero-dorsal tegmentum/pedunculo-pontine tegmentum (LDT/PPT), while the REM-OFF neurons are NA-ergic and largely located in the locus coeruleus (LC).[ It is important to note that some neurons in other parts of the brain have also been found to behave phenotypically as REM-OFF and REM-ON neurons; however, they have not been studied as extensively as those in LC and LDT/PPT. The LDT/PPT and LC neurons have reciprocal projections,[ which could be a direct connection or through GABA-ergic,[ glutamatergic or glycinergic[ interneurons. Additionally, these REMS-related neurons receive projections from widespread areas in the brain, including those involved in regulating sleep and waking (reviewed in[). Thus, it is evident that modulation of the activity of REMS-related neurons would be complex and so will be the regulation of REMS.

In an attempt to explain the neural regulation of REMS, based on single neuronal recording in freely moving normally behaving animals in isolated experiments, a simple model based on the Lotka-Volterra principle explaining the reciprocal interaction between REM-ON and REM-OFF neurons was proposed.[ The temporal relationship of reciprocal behavior of the REM-ON and REM-OFF neurons was confirmed by simultaneous recording of a pair of REM-ON and REM-OFF neuronal activities, along with waking–non-REMS (NREMS) and REMS in chronically prepared freely moving normally behaving cats.[ Subsequent extensive systematic in vivo and in vitro studies during the past two decades have confirmed the role of GABA-terminals and GABA-interneurons acting pre- and postsynaptically on REMS-related neurons[ for REMS regulation. These findings have been consolidated recently and a working model that activation of a de-activation process is responsible for REMS initiation and regulation has been proposed.[

Factors affecting LC–NA-ergic neurons modulate REMS and may induce REMS loss-related symptoms

The REMS is a multifactorial complex phenomenon; its regulation is also likely to be multi-dimensional. The role of many neurotransmitters and neuropeptides directly or indirectly have been implicated in the regulation of REMS.[ The LC is the primary site of NA-ergic neurons in the brain.[ These neurons normally cease activity during REMS;[ however, they do not cease activity during REMSD.[ On the other hand, if the LC–NA-ergic neurons were kept active for a short- or long-term by electrical stimulation[ or by preventing GABA-ergic inhibitory input to act on them[ or by increasing the Na–K ATPase activity on the LC neurons,[ REMS was reduced. Many factors that modulate REMS have been shown or proposed to mediate their action by modulating LC neurons. For example, normally, REMS does not appear during waking because wake–active[ areas including orexinergic perifornical area (PeF)[ activate, while the NREMS-areas[ in the brain inhibit[ the neurons in LC, the site of LC–REM-OFF neurons. Further, neuronal activity is modulated by microinjection of neurotransmitter agonist and antagonist into the LC-altered REMS.[ As the decrease in REMS is due to increased excitation and sustained activity of the LC– REM-OFF NA-ergic neurons, it is imperative that the later would elevate the NA level in the brain during REMSD and vice versa. Also, after REMSD, as tyrosine hydroxylase (TH) is increased and monoamine oxidase (MAO) is decreased, there would be increased NA in the brain.[ The possible mechanism of REMSD-associated elevated synaptic NA level at NA-ergic synapses and the brain at large is shown in Figure 1. Thus, in principle, either by modulating LC– NA-ergic neuronal activity, which in turn would control the quantity of NA-release, or by preventing the action of the elevated NA on potential target(s), REMS loss-associated symptoms may be ameliorated.

Figure 1

Noradrenalin (NA) concentration at the synapse is ultimately responsible for manifesting its effect. Its action is mediated by NA-level and its action on the postsynaptic adrenoceptors (ARs). However, the level of NA at the synapse is modulated by changes in either or combination of some or all of the following factors: (i) firing rate of NA-ergic neurons and (ii) changes in (a) TH activity, (b) NA transporter activity, (c) MAO activity and (d) number of α2-ARs and their activation of the NA-ergic neurons. Left panel (A) represents that if any one sub-type of receptor (e.g., by antagonist of α1-ARs as shown here) is blocked, relatively more NA becomes available to act on other sub-types of ARs, which may then express nonspecific side-effects (symptom). Although as an example we have shown that by blocking α1-ARs, a similar effect may be evident by blocking any other one type or combination of ARs types Right panel (B) represents action of NA on the pre-synaptic α2-ARs (autoreceptors) and thus modulating the release of NA

Modulation of effect of NA in the brain: A possible therapeutic approach to ameliorate REMS loss-associated symptoms

Normally cessation of NA-ergic REM-OFF neuronal activity is a prerequisite condition for generation and maintenance of REMS.[ If they do not cease activity, REMS does not appear and as a result of noncessation (i.e., continuation of activity) of LC neurons, there is likely to be elevated level of NA in the brain. This elevated NA in the brain in turn has been reported to induce many, if not most, of the REMS loss-associated symptoms.[ These propositions may be supported by the fact that many of the diseases associated with REMS loss[ show an elevated level of NA or express symptoms that could be due to elevated level of NA.[ Therefore, effective neutralization of the action of released NA, particularly at the site of its release/action, could be a natural and preferred choice of therapeutic intervention to counter the REMS loss-induced effects and associated symptoms. This could be achieved by either or combination of the following:

Blocking the postsynaptic AR to prevent the action of NA;

By elevating the level of monoamine oxidase that breaks down the released NA;

By increasing the NA-transporter and reducing the effective concentration of NA at the synaptic sites;

By modulating the synthesis and release of NA;

By influencing inputs on the LC–NA-ergic neurons thereby altering their activities and release of NA; or

By targeting the presynaptic ARs at the NA-ergic terminal and thus modulating the release of NA

Various possibilities targeting the release of NA at the terminal, along with their advantages and disadvantages, have been discussed in the following sections considering each one of them as a potential therapeutic approach for the treatment of REMSD-associated symptoms.

NA-ergic AR

NA acts through ARs that belong to G-protein-coupled receptor and may give rise to a variety of responses. There are mainly three subtypes of ARs: α1, α2 and β-ARs; they share a high degree of amino acid homology, especially within the ligand binding pockets around the transmembrane regions.[ Ligand-binding residues confer subtype-specific selectivity to a particular receptor. α1-ARs are further sub-classified as α1A, α1B and α1D types and are coupled to the Gq/11 family.[ Stimulation of these α1 ARs recruits signaling pathways involving activation of phospholipase C, D, A2 and MAP kinases and, subsequently, modulate transcriptional activation of early and late response genes. For example, it has been reported to affect the expression and activation of Ca2+ channels, Na+/H+ exchangers and K+ channels,[ which in turn are reported to modulate neuronal functions. At least in neurons, NA has been reported to close L-type Ca2+channels by acting on α1-AR.[

The subtypes of β-ARs, viz. β1, β2 and β3, activate effectors by coupling to Gs or Gi molecules in a time-and dose-dependent manner.[ Although the action of NA on different types of cells might not have been studied, at least in cardiomyocytes, stimulation of β-AR by nonselective agonist results in Gs-mediated enhanced cAMP generation and activation of the related downstream events. The β-AR-mediated effects enhance contractility of the cardiac muscles either by one or in combination with increasing Ca2+ influx through L-type Ca2+ channels, increased reuptake of Ca2+ by disinhibition of sarcoplasmic reticular Ca2+-ATPase and modulation of myofilament Ca2+ sensitivity.[ The β2-ARs can also couple to Gi protein to inhibit the downstream effectors.[

The α2-AR family has α2a, α2b and α2c receptor subtypes, and they are encoded by three distinct genes. The α2-AR couples to the Gi subunit and is involved in the modulation (usually inhibition) of NA release.[ Activation of Gi leads to inhibition of adenylate cyclase, which results in decreased cAMP generation. Coupling to several other signaling pathways has also been reported upon activation of α2-ARs for modulation of neurotransmitter functions. These include activation of K+ channels, inhibition of Ca2+ channels, activation of Na+/H+ antiporter, mobilization of intracellular Ca2+ and activation of the mitogen-activated protein kinase (MAPK) cascade.[

Targeting the postsynaptic α1-ARs

α1-ARs are distributed on the neurons throughout the brain, including those in the areas responsible for REMS regulation as such as well as on neurons responsible for functions related to REMS. Presence of α1-ARs on the wake–active neurons in the pedunculo-pontine tegmentum (PPT)[ and on thermosensitive neurons in the preoptic area[ have been reported, which in principle may mediate REMSD-associated changes in brain functions. α1-AR antagonists like prazosin (PRZ) have been widely used in the treatment of sleep (including REMS) disturbance, hypertension, posttraumatic stress disorder (PTSD) and anxiety.[ Methoxamine, an α1-AR agonist, increased wakefulness while decreasing NREMS and REMS,[ which was reversed by PRZ. In stress-sensitive WKY (Wistar Kyoto) rats, REMS was fragmented by electric shock, and this was prevented by PRZ.[ Similarly, in lower vertebrate model using zebrafish, PRZ treatment reduced the effect of sleep deprivation on anxiety.[ REMSD caused many changes, including cellular morphology and biochemical parameters, and molecular expressions of Na–K ATPase have been shown to recover by treatment with PRZ.[ These studies support that blocking α1-AR activation could reduce the REMS loss-associated symptoms. Mignot et al.[ investigated the role of central α1-ARs in cataplexy in genetically modified narcoleptic Doberman pinschers. Treatment of narcoleptic dogs with PRZ exacerbated cataplexy, whereas treatment with α1-agonist, methoxamine, ameliorated it. However, there are a few conflicting reports as well. For example, oral administration of PRZ shortened quiet waking and REMS but increased active waking and slow wave sleep,[ whereas other α1-AR antagonists, thymoxamine and mesoridazine increased REMS in humans.[ Thus, the conditions induced by sleep loss, including REMS-loss, were rescued/recovered by activating/deactivating (as the case may be) the α1-ARs supporting our contention. These findings suggest that if NA was prevented to act on α1-ARs, REMS was increased and the REMS loss-associated symptoms were decreased.[

Targeting postsynaptic β-AR

Improvement in sleep disturbance-associated mood disorder, severe depression and sleep apnea have been reported by treatment with β-AR blockers in both animals and humans. [ Also, REMSD-associated aggressiveness has been suggested to be mediated by NA acting on β-AR.[ β-agonist, isoproterenol, suppressed while β-antagonist, propranolol, consistently enhanced REMS episodes.[ Although the amelioration of REMS loss-associated symptoms could be due to improved REMS, their causal and temporal relationship need validation. Microinjection of isoproterenol into the medial septal region of the basal forebrain significantly increased the time spent in wakefulness, while there was a near-complete suppression of REMS.[ Based on these findings, it has been postulated that NA released from the LC REM-OFF neurons acts on β-ARs present on REM-ON neurons and inhibits those preventing REMS generation.[ As a corollary, it has been suggested that β-blockers would withdraw the NA-induced inhibition of the REM-ON neurons and facilitate their activation, leading to the generation of REMS; however, there are differences in specificity of chemicals acting on receptors. For example, Kostis and Rosen (1987)[ reported that hydrophilic β-blocker (e.g., atenolol) do not affect sleep, while lipophilic β-blocker (e.g., pindolol) disturbs sleep continuity. Similarly, in another study, β-AR blockers, acebutolol and metoprolol, did not show any effect on sleep pattern.[ Acute administration of β3-AR agonist, CL316243 reduced while its prolonged treatment did not affect REMS.[

Thus, although it has been well established that the adrenergic system plays a significant role in REMS regulation, the precise mechanism of how it (β-blockers in particular) mediates the action needs further study. It may be noted that these drugs are widely used in cases of cardiac arrhythmias and hypertension; however, their associated undesirable side-effects of sleep disturbances and insomnia limits their use.[ Hence, it is important to understand the use of these chemicals as drugs possibly after classifying patients so that these drugs may be used in combination with other molecules to reduce the side-effects. In addition to the application of agonist and antagonist mentioned above, molecules activating reuptake and/or breakdown of NA by MAO-A [Figure 1] at the synaptic terminal also have been used in the treatment of sleep disturbances.[ In this strategy also, ultimately, the effective level of NA at the synaptic cleft is altered.

Targeting the release of NA

Like any other neurotransmitter, NA is released from the presynaptic terminals and it acts on the pre- and postsynaptic receptors for inducing its action. Thus, all other conditions remaining unchanged, the quantity of NA at the synaptic site decides its effect on expressing a behavior. We have discussed above how the effects of the released NA (due to loss of REMS) could be reduced by directly inactivating (NA degradation or re-uptake) the released NA or by preventing the released NA to act on the postsynaptic receptors. Another option of modulating the action of a neurotransmitter at the synaptic site is to modulate the release of neurotransmitter per se (NA at this instance) from the presynaptic site, which has also been used in treatment as well as in research.[

Here, we propose that manipulation of auto-regulatory mechanism of release of neurotransmitter than preventing the action of already released NA possibly would be a better option to maintain the effective level of NA at the synaptic site. This is because use of antagonist of one subtype of AR, although preventing the action of NA on such receptor subtypes, the already released NA remains or becomes available to act on other subtypes of ARs leading to added complications. This is because, for example, under normal conditions, the released quantity of NA will be such that it acts on an optimum number of one or more subtypes of ARs to get a function expressed at an optimum level. However, if a subtype of AR is blocked by its antagonist, the equilibrium of the ligand (NA in this case) to subtypes of receptor (ARs in this instance) would shift and, although a function may get modulated, possibly toward a desired level, other functions may get affected or get biased, expressing undesirable side-effects possibly as compensatory effect(s). This could be due to various reasons including overcoming the threshold of activation of another subtype of ARs, which was not significantly affected under normal condition, under the condition of blocking of one subtype of AR. After blocking one subtype of AR, the available NA might reach the threshold of activation of one or more of other subtypes of ARs expressing undesirable side-effects. This view may be supported by the fact that NA acting on different subtypes of ARs in the same brain area, viz. preoptic area, modulates sleeping-, waking- and thermo-regulation.[ The literature on experimental as well as clinical practices mentions many such side-effects or associated phenomena that are often ignored due to lack of recorded consistent effects. Although in all cases we do not know the cause and effect relationship, our reasoning offers explanation for the same, which is justifiable and testable. Hence, we propose that in order to target REMS loss-associated induced symptoms, a more effective strategy could be to target modulation of the release of NA itself. However, the limitation is how to design site-specific moderation of release of NA.

Targeting the presynaptic α2 auto receptors

There are many examples where patho-physiologial processes have been reported to be modulated by targeting the α2-ARs. Initially, it was believed that α2-ARs are present exclusively on the presynaptic site; however, recent reports suggest that in addition to the presynaptic site, they are present on the postsynaptic site as well.[ The unique property of α2-AR being present on the presynaptic site is that excess NA at the terminal (synaptic cleft) activates these α2-ARs, preventing further release of NA; thus, the level of NA is auto-regulated at the terminal. The released NA inhibits the influx of calcium ions and thereby prevents further release of NA.[ As there are collateral feedback inputs on the NA-ergic neurons in the LC,[ it is likely that the collateral inputs on to itself and rhythmic release of NA due to auto-regulation of NA maintains the rhythmic firing of the LC–REM-OFF neurons and prevents appearance of REMS, especially during waking.[

The α2-AR agonist, clonidine, has often been used as the treatment to reduce NA release. For example, clonidine and its analogue has been used in treating patient suffering from Crisponi syndrome.[ In humans, a lower dose of clonidine increased while a higher dose decreased REMS.[ The findings suggested that before assigning a therapeutic role to an agent, its dose, time and secondary or associated physiological impact should be evaluated. Dexmedetomidine, a more selective α2-AR agonist than clonidine, has been reported to decrease sleep and reduce c-Fos expression in the LC neurons.[ The spindles observed under sedation with dexmedetomidine are qualitatively similar to those during natural sleep. However, it has associated problems, e.g. it works even when NA level is severely depleted by various other agents like reserpine and DSP-4.[ Yohimbine (3 mg/kg), an α2-AR antagonist, augmented waking and reduced sleep.[ Activation of α2-ARs in and around LC reduced NA release and, consequently, reduced NA-mediated effects in the central nervous system.[ α2-ARs have been shown to be involved in REMS-associated thermoregulatory responses.[ REMSD-induced increased NA was responsible for impairment in learning and memory in rats, which was significantly improved by treating with α2-AR agonists.[

Proposed model for the treatment of REMSD-associated symptoms

In the brain, the LC is the primary site of NA-ergic neurons. As these neurons project throughout the brain and are responsible for most of NA in the brain, alterations in the activity of these LC–NA-ergic neurons ultimately modulate the level of NA in the brain. These neurons normally cease firing during REMS; however, they do not stop firing upon REMSD. Therefore, during REMS, although normally NA is washed-off from the NA-ergic projected sites, up on REMSD those sites have an increased level of NA. The LC neuronal activities are affected by the inputs they receive from within the brain as well as from the periphery; also, they are influenced by many other physico-chemical factors. Thus, it is understandable that the REMS is affected by wide varieties of inputs originating from within or from external inputs and, upon REMS disturbance, there will be changes in the level of NA in different parts of the brain. This explanation supports the observation that REMS is affected (may be even secondarily) in almost all altered states, e.g. acute fever to chronic and complex psycho-somatic disorders.

Although until now we do not know the exact details, REMS disturbance appears to be a primary cause of many acute to chronic disorders and, depending on the specific neurons that are affected and their projections in the brain, the level of NA would be modulated. This altered NA level would then modulate the physiological activities, which then get expressed as REMSD-associated symptoms. In other case(s) where the level of NA in the brain is elevated as a primary cause, the REMS-loss may be induced as a secondary effect and the associated pathological symptoms are expressed depending on the susceptibility of the neurons affected and their projections where the NA level is increased. Further, as most NA in the brain is released from the LC neurons, which is generally of the REM-OFF type, the REMS is also affected as a secondary/associated effect/response. It is possible that initial change in the level of NA due to REMS-loss could be a compensatory effect, which is often beneficial. However, if the loss is continued, it leads to chronic disorder and the mechanism needs to be understood in detail.

Based on the arguments given above, we propose that diseases associated with REMS loss need to be treated by a combination of factor(s), α2-ARs agonist, which would reduce the release of NA along with agonist or antagonist of such receptor, which presumably is the primary cause of modulating the LC neurons. It may be a tough call on how to decide on the latter; however, we think that it could be done symptomatically on the basis of expressed symptoms. Although apparently it may be difficult to conceive at present, at least it may be said with certainty that all sleep/REMS disorder patient cannot be treated in the same manner (by the same drug) and, if this is done, including due to self-medication, it would create more complications as is often experienced; however, which might not have been reported as frequently due to a lack of consistent expression of symptoms. Finally, we propose that modulation of NA-release is likely to be the desirable treatment for REMS loss-associated symptoms. This may be achieved by modulating the firing rate of the LC–NA-ergic neurons in combination with or without independently modulating the presynaptic α2-ARs; however, the challenge is to design such a targeted delivery system to be used as a therapy.

Basic principle of such targeted drug delivery system as a proof of principle

As discussed above, in principle, delivery of α2-AR agonist to a desired site is the likely solution. However, the challenge is how to deliver the drug to a desired target, e.g. where the NA-ergic inputs are on the postsynaptic neurons possessing specific subtypes of receptors (cholinergic, GABA-ergic, etc. for instance). Subject to experimental verification, we suggest (propose) that the desired α2-AR agonist could be packaged in liposomes having affinity for the target site where the drug (the cargo) needs to be delivered; the affinity of the designed molecules on the liposomes could be made of various combinations. Once such designed liposomes (the vehicles) are targeted at specific sites, the drug (the cargo) will be delivered at the desired site; the vehicle may be designed as per need.

CONCLUSION

REMS is a complex behavioral phenomenon and its loss affects various psycho–patho-physio-logical processes. NA level is elevated in the brain upon REMS-loss, which has been reported to be responsible for inducing several REMS loss-associated symptoms. We propose that reduction of NA-release either by modulating the NA-ergic neuronal firing rate alone or in combination with activating the presynaptic α2-ARs is likely to be preferred and promising strategies for treating REMS loss-associated symptoms. Subject to confirmation, as a proof of principle, we have also proposed a possible approach for targeted drug delivery.