Invited Symposium: Role of the Basal Forebrain Neurons in Cortical Activation and Behavioural State Regulation
Numerous experiments, using a variety of behavioral paradigms for the assessment of different aspects of attention, have concluded that the integrity of cortical cholinergic inputs is essential for the detection, selection and processing of stimuli and associations [1-5]. Furthermore, basal forebrain neuropharmacological manipulations known to increase or decrease the excitability of cortical cholinergic inputs in intact rats bidirectionally alter the attentional abilities of rats assessed in tasks designed for the measurement of sustained or divided attention [6,7]. Electrophysiological studies demonstrated that the increases in firing rate in medial prefrontal neurons in result to the presentation of a visual distractor to rats performing a sustained attention task depend on the integrity of cholinergic inputs to this region . These and other data have strongly supported hypotheses about the specific attentional functions mediated via cortical cholinergic inputs.
The hypothesis that cortical cholinergic inputs mediate attentional abilities appears to be in conflict with the more traditional notions that the basal forebrain corticopetal projection represent a rostral extension of the ascending "arousal" systems . Below, the evidence in support of "arousal"-like functions of cortical cholinergic inputs will be briefly discussed, and then alternative functional descriptions of the converging brainstem, telencephalic and limbic afferents to basal forebrain cholinergic neurons will be suggested on the basis of recent behavioral and neuropharmacological data.
2. The construct "arousal"
The attribution of "arousal" functions to basal forebrain corticopetal neurons has been derived from pharmacological evidence in support of a "cholinergic nature" of cortical arousal, specifically from studies showing relationships between EEG desynchronization, increases in spontaneous alertness, and increases in cortical acetylcholine (ACh) turnover or cholinergic receptor stimulation . This perspective has further corroborated notions about the basal forebrain corticopetal system representing a rostral extension of the ascending reticular activating system, as evidenced by the "reticular" anatomical and morphological characteristics of basal forebrain corticopetal projections and the brainstem projections to basal forebrain neurons (e.g., ). Presently, three main lines of evidence continue to substantiate the role of basal forebrain corticopetal projections in the regulation of "arousal". First, manipulations of the excitability of basal forebrain neurons modify cortical event-related potentials (e.g., ) and other electroencephalographic (EEG) measures (e.g., [13-13]). Second, neuronal activity in the basal forebrain correlates with EEG activation (e.g., [16-18]). Third, basal forebrain neurons are involved in the diurnal regulation of sleep parameters and associated EEG activity, probably via connections with midbrain reticular and pontine structures that are part of the ascending reticular system [19-23]. Thus, the attribution of arousal functions to basal forebrain corticopetal projections has remained largely driven by research linking the basal forebrain with brainstem ascending systems and their "arousal"-mediating functions, and by studies assessing EEG variables and global behavioral states.
A wide range of rather generally defined transitions from sleep or unconscious states to wakefulness or conscious awareness or the effective cortical processing of information have been collapsed into the construct "arousal" (e.g., [24,11]. While it is generally unlikely that such a unitary construct continues to assist in attempts to identify and dissociate specific functions mediated via ascending systems, particularly in light of more contemporary evidence revealing their unexpectedly complex anatomical organization (e.g., [25-27], Robbins and Everitt  stress that even the available empirical evidence render the arousal construct untenable. In particular, they stress the lack of intercorrelations between various indices of arousal that would be expected from such a unitary construct. The following discussion on the modulation of the cortical ACh-mediated attentional functions by brainstem afferent projections to basal forebrain neurons attempts to approach an alternative conceptualization, by stressing the precise nature of the modulation of forebrain information processing via brainstem afferents, and by determining the conditions which allow brainstem ascending projections to alter the functions of their basal forebrain target networks.
3. Functions of brainstem afferents
As briefly summarized above, cortical cholinergic inputs mediate attentional functions, ranging from aspects of sustained and selective attention to divided attention and the regulation of processing capacity or the allocation of processing resources. It is important to stress the fact that these attentional functions are precisely defined, and that the majority of the available experiments employed paradigms with solid construct validity. As has been discussed elsewhere, important tenants of this hypothesis suggest that changes in the activity in cortical cholinergic inputs are not cortical area-specific (e.g., [29,30,2], and that the selectivity of the behavioral effects of cortical ACh are based on close temporal interactions with converging sensory or associational cortical inputs (e.g., ). In addition to brainstem afferents of basal forebrain neurons which will be focused on below, major telencephalic projections, including afferents from the extended amygdala converge on basal forebrain neurons [32,33] and modulate the excitability of basal forebrain corticopetal neurons in the context of attention-demanding situations [2,34]. Our previous research has focused on the regulation of cortical ACh efflux and associated attentional functions by GABAergic inputs to basal forebrain cholinergic neurons, presumably originating in the nucleus accumbens (for review see [35,2,36,7], and by glutamatergic inputs from telencephalic areas [37,38]. The data from these experiments have strongly supported the hypothesis that the attentional functions of basal forebrain corticopetal cholinergic neurons primarily are modulated by such telencephalic inputs to the basal forebrain. As will be discussed next, the available data from studies on the role of noradrenergic projections to the basal forebrain in the regulation of cortical ACh efflux and attention may illustrate the ability of brainstem ascending projections to basal forebrain neurons to modulate telencephalic networks and thus, regulate the excitability of these neurons and the associated attentional functions.
Noradrenergic projections to basal forebrain cholinergic neurons originate in the locus coeruleus and the A5 group in the ventrolateral tegmentum (e.g., ). The cholinergic neurons in the basal forebrain receive diffuse noradrenergic inputs, with neurons in the caudal substantia innominata showing a particularly dense noradrenergic input [40,41]. Basal forebrain cholinergic neurons are predominantly depolarized via a1 receptors [42,43].
While the performance of rats in a sustained attention task firmly depends on the integrity of basal forebrain corticopetal cholinergic neurons (e.g., ) and on the GABAergic  or glutamatergic innervation of these neurons from telencephalic areas , lesions of the dorsal noradrenergic bundle (DNB) which decreased forebrain noradrenaline contents by more than 90 % did not affect performance in this task . This finding corresponds with results from the effects of similar lesions on the performance of rats in the 5-choice serial reaction time task [46,47]. While noradrenergic inputs may play no role in the mediation of the visual attention performance of well-trained animals, the performance effects of distractor presented in a different modality in DNB lesioned rats  point to the possibility that the available concepts of attention require incorporation of the effects of stressors to provide an interpretative frameworks for the complex, task-parameter dependent effects of DNB lesions .
As discussed by Aston-Jones et al. [49,50], the major afferent innervation of the LC originates from the nuclei paragigantocellularis (PGi) and prepositus hypoglossi (PrH). Increases in activity in the LC is driven primarily by the projections from the PGi which itself mediates sympathoexcitatory mechanisms. Aston-Jones and co-workers suggest that the attentional effects of immediately relevant stimuli activate the LC primarily via the PGi and thus, the ascending noradrenergic system mediates the attentional components of the response under such conditions. This abbreviated discussion of their hypothesis would predict that lesions of the noradrenergic bundle may not affect performance in our sustained attention task, as it does not involve sufficiently provocative stimuli. In contrast, the performance effects of a stressor observed in the experiments by Robbins and co-workers (see above) may be associated with the necessary sympathetic activation to affect LC activity which, in lesioned animals, reveals the consequences of a dysfunctional ascending noradrenergic system. Thus, the functional significance of basal forebrain noradrenergic-cholinergic interactions can be, at least hypothetically, described: sustained attentional performance necessarily requires the integrity of the cortical cholinergic afferents but not noradrenergic cortical afferents. Optimization of the subjects' attentional processing in response to urgent (e.g., aversive) stimuli is mediated via noradrenergic mechanisms which acts in the basal forebrain to foster its recruitment by telencephalic afferent circuits.
This hypothesis gains further support by studies on the effects of infusions of noradrenergic drugs into the basal forebrain on cortical ACh efflux . These data suggest that infusion of the a1 agonist phenylephrine into the basal forebrain of na´ve rats is sufficient to allow co-infusions of NMDA to yield increases in cortical ACh efflux. Infusions of either phenylephrine or NMDA did not affect cortical ACh efflux. However, a different scenario arises in animals in which cortical ACh efflux is already activated by the presentation of a conditioned stimulus for palatable food (e.g., ) or in which NMDA augmented the increase in cortical ACh efflux produced by such a conditioned stimulus . Infusion of the a1 antagonist prazosin into the basal forebrain did not affect the increases in ACh efflux that resulted form the presentation of the conditioned stimulus. Moreover, co-perfusion of prazosin did not affect the NMDA-induced augmentation of ACh efflux . These data suggest that noradrenergic stimulation of basal forebrain cholinergic neurons contribute to the ability of telencephalic inputs to "recruit" the basal forebrain corticopetal system, but that these brainstem inputs do not further modulate the excitability of this system once it is recruited by telencephalic inputs.
These basal forebrain noradrenergic-cholinergic interactions have been speculated to play a major role in disorders characterized by the stress-induced processing of selected classes of stimuli and associations, such as the biased processing of anxiety-related stimuli and contexts in subjects suffering from generalized anxiety disorder or panic disorder. Based on the innervation of the locus coeruleus (LC) by the PGi and PrH (see above), Aston-Jones and colleagues proposed a role of LC efferents in emotional "arousal" or "activation" . Specifically, anxiety states are associated with increases in autonomic reactivity that are mediated via descending projections from amygdaloid and other structures traditionally considered to mediate anxiety-like behaviors (see Fig. 4 in ). Increases in autonomic reactivity are monitored by the LC via its afferent projections. The resulting stimulation of basal forebrain cholinergic neurons via LC ascending projections is hypothesized to further bias cortical information processing toward anxiety-related stimuli and associations (see the discussion in ; see also ).
The available data on noradrenergic-basal forebrain cholinergic interactions provide a preliminary basis for a specific functional description of the role of ascending noradrenergic inputs to the basal forebrain. Noradrenergic stimulation is hypothesized to contribute to the ability of telencenphalic afferents of the basal forebrain to recruit corticopetal projections and thereby to initiate attentional information processing.
The results from studies focusing in the role of basal forebrain neurons in the regulation of sleep, particularly REM sleep, may further illuminate the role of brainstem inputs in the regulation of basal forebrain corticopetal neurons and attentional functions. High discharge rates of basal forebrain neurons during REM sleep have been reported , and they may, at least in part, be due to cholinergic stimulation via afferents originating in the pons . These data invite the speculation that the dissociated selection and processing of associations in REM sleep is driven, at least in part, by brainstem input-mediated stimulation of basal forebrain corticopetal neurons (see also [55,56]. As cortical ACh in awake animals interacts with other converging inputs to augment the processing of sensory and associational information (e.g., [57,58,2], and as it may be speculated that during REM sleep, brainstem inputs provide the functionally dominating input to basal forebrain neurons - as opposed to their modulatory interactions with telencephalic inputs to basal forebrain neurons mediating awake, attentional processing - this "de-telencephalic" activation of the basal forebrain during REM sleep may mediate the random selection of associations and/or the attempts to integrate such associations . Moreover, the long-term, plastic cortical effects resulting from the cortical convergence of cholinergic receptor stimulation and other sensory or associational input (e.g., [30,60,57]) may mediate the facilitation of stimulus selection and processing during learning and rehearsal-induced facilitation of associational processes; such a coordinated pairing of cortical inputs may not occur during REM sleep, thus dramatically limiting the recall for the materials processed during REM sleep, thus also limiting the potential that REM-associated processing infiltrates the cognitive integrity of the awake subject.
While the cognitive functions of basal forebrain corticopetal projections are increasingly well defined, it is obvious from these speculations that research on the neuropharmacological interactions between basal forebrain telencephalic afferents and brainstem afferents and on their behavioral/cognitive significance is in dire need. Studies on such interactions are very likely to yield hypothesis that have no use for the broad construct "arousal" but conceivably describe such interactions in specific cognitive terms, including the cognitive consequences of pathological alterations in the innervation of basal forebrain neurons.
The authors' research was supported by NIH Grants NS32938, MH57436, NS37026 and MH01072.
| Discussion Board | Previous Page | Your Symposium |
|Sarter, M; Bruno, JP; (1998). Cortical Afferents Originating in the Basal Forebrain: Mediation of Specific Aspects of Attentional Processing Versus General Assumptions about Cortical Activation and Behavioral State.. Presented at INABIS '98 - 5th Internet World Congress on Biomedical Sciences at McMaster University, Canada, Dec 7-16th. Invited Symposium. Available at URL http://www.mcmaster.ca/inabis98/semba/sarter0163/index.html|
|© 1998 Author(s) Hold Copyright|