Cholinergic projections and receptors in V1
Depending on the types of target neurons, variations in postsynaptic receptor subtypes, and the different subcellular sites targeted, ACh have been shown to have diverse effects upon visual cortical neurons.
Early studies documented the laminar distribution pattern of cholinergic axons throughout the cortex based on electron microscopic visualization of choline acetyl transferase (ChAT) immunoreactive synapses (Mesulam et al., 1983; Houser et al., 1985). ChAT catalyzes ACh synthesis using acetyl CoA and choline as substrates. Given that ChAT is the acetylcholine-synthesizing enzyme, it serves as a marker for cholinergic neurons. The cerebral cortex has been found to have a diffuse cholinergic innervation, mostly from neurons in the nucleus Basilis of Meynert of the basal forebrain (Mesulam et al., 1983; Rye et al., 1984). In V1, ChAT immunoreactive fiber density varies across species. Humans show the largest number of immunoreactive fibers in layers I-IIIA (Mesulam et al., 1992), while the highest levels found in the adult macaque monkey are in layers I and IV (Mrzljak and Goldman-Rakic, 1993). ChAT immunoreactive fiber density also shows unique species-specific patterns in V1 for the rat (Parnavelas et al., 1986), mouse (Kitt et al., 1994), cat (Stichel and Singer, 1987a, b), and ferret (Henderson, 1987). For example, in rats and mice, although ChAT-positive fibers were distributed throughout all layers, highest densities were detected in cortical layers IV and V. Conversely, in the cat V1, highest fiber density was detected for layer I, while low numbers of ChAT immunolabeled fibers were found in the infragranular layers (Parnavelas et al., 1986; Stichel and Singer, 1987a; Kitt et al., 1994). It is therefore clear that there is considerable variability in the distribution patterns of cholinergic fibers, however, it is not currently known if and how these disparate patterns might contribute to species-specific visual processing.
Both muscarinic and nicotinic ACh receptor classes are found in the visual cortex (Zilles et al., 1989). Some muscarinic subtypes show a similar laminar distribution across V1 in several species, while the distribution pattern of other subtypes differs in a species-specific manner (Gu, 2003). Of the mus-carinic receptors, anatomical localization in the cortex is best described for M1 and M2 receptor subtypes, largely because there are no specific labels for other subtypes (Gu, 2002). The M1 muscarinic receptor, for example, is most dense in supragranular layers and least in middle cortical layers of the rat (Schliebs et al., 1989; Zilles et al., 1989), cat (Prusky and Cynader, 1990), and primates (Lidow et al., 1989a; Tigges et al., 1997). The M2 subtype distribution, on the other hand, differs widely across these species. Even across individuals of the same species, M2 receptor distribution can vary greatly. Peak density occurs in superficial layers and in lower layers IVc and IVa of the rhesus monkey (Lidow et al., 1989a; Tigges et al., 1997). In the cat V1, M2 receptors are highly concentrated in layers I and V, although significant levels of this receptor are found in both supra- and infragranular layers (Prusky and Cynader, 1990). In the rat, although M2 expression was found throughout all cortical layers, highest levels were found in the thalamorecipient layer IV, while lowest expression was detected in supragranular layers (Rossner et al., 1994).
Unlike muscarinic receptors, which are coupled to G proteins and show a long latency of action, all nicotinic ACh receptors (nAChRs) belong to the superfamily of ligand-gated ion channels. Most cortical nAChRs are thought to exist as heteromers of alpha4 and beta2 subunits or as alpha7 homomers, and are found throughout the cortex (Metherate, 2004). The nAChRs can also be classified according to their affinity for alpha-bungarotoxin: low affinity nAChRs are alpha-bungarotoxin sensitive and high affinity nAChRs are not (reviewed in Gotti and Clementi, 2004).
These different nAChR subtypes not only serve distinct functions, but they show a different cortical distribution. Low-affinity nAChRs occur mainly within supragranular and infragranular layers, while high-affinity nAChRs are densest in the middle cortical layers (Clarke et al., 1985). Some nAChRs are located on thalamocortical afferent terminals primarily within cortical layer IV, but most receptors appear to be postsynaptic (Prusky et al., 1987; Parkinson et al., 1988b; Lavine et al., 1997). Most nAChRs appear on pyramidal cells and spiny stellate cells and, at lower frequencies, on inhibitory interneurons. Subcellular localization on pyramidal cells emphasized receptor placement on cell bodies and apical dendrites (Zilles et al., 1989).
Cholinergic effects on physiological properties of V1 neurons and plasticity
Like studies on the noradrenergic system, the effects of acetylcholine (ACh) on the properties of V1 neurons have primarily been investigated using chemical lesioning of cholinergic neurons, iontophoretic application of ACh and delivery of selective antagonists for particular types of cholinergic receptors. As detailed above, cholinergic basal forebrain neurons innervate the mammalian visual cortex profusely, suggesting a putative role for ACh in modulating visually-driven response properties in V1 (Saper, 1984; Mesulam et al., 1992). Studies conducted in cat primary visual cortex with electrical (via lateral geniculate nucleus stimulation) and visual stimuli have directly investigated the role of ACh in V1 physiology.
Local ACh application triggers both facilitation as well as depression of neuronal responses, results that are comparable at some levels with those obtained with noradrenaline (Sillito and Kemp, 1983; Sato et al., 1987a). Facilitatory effects however are more prevalent when compared with AChmediated suppression (Sillito and Kemp, 1983; Sato et al., 1987a).
Local ACh application has been reported to markedly influence V1 response properties, including enhanced directional and orientation selectivity (Sillito and Kemp, 1983). Unlike GABAergic (discussed in Chapter 11), and to a lesser degree, noradrenergic influences, cholinergic influences do not appear to regulate receptive field (RF) area, suggesting that this neurotrans-mitter does not, to any degree, determine the functional tuning of RFs. Furthermore, no differential actions of ACh on simple versus complex cells have been observed (Sillito and Kemp, 1983).
Another approach that has proven useful in investigating the role of ACh in visual processing has been the use of cytotoxic lesions directed at sites involved in ACh synthesis. For example, Sato and colleagues (1987b) conducted kainic-acid-induced unilateral lesions in the cat nucleus basalis magnocellularis (NBM), the major source of ACh projections to the visual cortex. Neurons located ipsilaterally to the lesion appeared to have undergone a substantial decrease in visually-evoked responses, while contralateral neurons were unaffected. Interestingly, direction and orientation selectivity were unaffected by depletion of cholinergic neurons in the NBM. However, rescue of normal cholinergic transmission for the ipsilateral side by local ACh application triggered both the response of originally silent neurons, and the facilitation of neuronal responses in the large majority of cells recorded (Sato et al., 1987b).
It has been reported that the effects of ACh on V1 neurons are blocked by infusion of pharmacological agents directed at the muscarinic (but not nico-tinic) cholinergic receptors (Sato et al., 1987a), suggesting that ACh-mediated facilitation might be correlated with disinhibition. This effect was likely mediated through a reduction of the potassium current. In experiments conducted by another group (Sillito and Kemp, 1983), where the facilitatory effects of ACh were compared to disinhibition triggered by bicuculline application, it was concluded that the effects of both neurotransmitter systems are dissimilar.
As reported for the noradrenergic system, the effects of ACh-mediated facilitation or suppression appear to be layer-specific, with neuronal response suppression predominantly occurring for cells located in the granular cell layer and layer III, while facilitation was detected across all cortical layers (Sillito and Kemp, 1983; Sato et al., 1987a). This differential impact of ACh on V1 neurons might be correlated with the asymmetric expression of different receptor subtypes in sub-populations of visual cortical neurons. In fact, it has been shown that M1 receptors are more prevalent in the granular and supragranular layers of V1 across a number of experimental models (Lidow et al., 1989a; Schliebs et al., 1989; Zilles et al., 1989; Prusky and Cynader, 1990). The highest concentration of M2 receptors, however, appears to be regulated in a species-specific manner (Lidow et al., 1989a; Zilles et al., 1989; Prusky and Cynader, 1990). To our knowledge, no systematic analysis of the contributions of each receptor subtype to either facilitation or suppression behavior has been conducted to date.
A classic paradigm for postnatal visual system plasticity involves shifts in ocular dominance induced by monocular deprivation (reviewed in Miller, 1994; Bear and Rittenhouse, 1999; Katz and Crowley, 2002). The contributions of cholinergic transmission to this form of plasticity were originally investigated by Bear and Singer (1986), who used excitotoxic lesions directed at the cholinergic basal forebrain of kittens. These authors observed that by selectively targeting either the cholinergic or the noradrenergic system (by infusion of 6-hydroxy-
dopamine; 6-OHDA), shifts of ocular dominance were preserved. Conversely, dual ablation of ACh and NA transmission severely delayed this form of plasticity, suggesting a synergistic effect between acetylcholine and noradrenaline enabling this form of visual cortical plasticity (Bear and Singer, 1986). Subsequent experiments using this same plasticity paradigm coupled with local infusion of specific cholinergic receptor antagonists elegantly showed that ACh plays a direct role in ocular dominance plasticity, mediated through muscarinic M1 receptors (Gu and Singer, 1993). Antagonism of nicotinic ACh receptors does not seem to play a role in ocular dominance shifts but rather has been shown to enhance visually-driven responses in V1 (Parkinson et al., 1988b; Gu and Singer, 1993). As discussed above, M1 receptor activation has been proposed to participate in visual cortical plasticity by enhancing activity via a decrease in potassium conductance. Another alternative for M1 participation in the plasticity response is in the process that leads to the cleavage of phosphatidylinositol biphosphate (PIP2), a membrane phospholipid that is an important component of several signal transduction pathways. PIP2 cleavage leads to the generation of two products, diacylglycerol (DAG; the polar domain of the original PIP2 structure) and inositol triphosphate (IP3; the apolar domain) (Berridge, 1998). Given that IP3 is apolar, it diffuses into the cytoplasm where it can then bind to specific IP3 receptors in the endoplasmic reticulum, a major source of intracellular calcium. Mobilized calcium from intracellular stores appears to directly increase NMDA-receptor conductance in hippocampal neurons (Markram and Segal, 1992). NMDA receptors are repeatedly implicated in a number of experience-dependent synaptic modifications (for a review see Collingridge and Singer, 1990). Finally, ACh is also suggested to increase the signal-to-noise ratio for processing within V1, given that its iontophoretic application does not lead to enhancement of spontaneous neuronal activity (Sato et al., 1987a).
A recent study has demonstrated that pharmacological intervention directed at the cholinergic system within cat V1 affects stimulus-driven gamma oscillations and the associated neuronal synchronization via muscarinic receptors (Rodriguez et al., 2004). In addition, ACh agonist application along with the presentation of visual stimuli that are known to trigger synchronization lead to long-lasting enhancement of gamma oscillations and synchrony in visual cortical neurons (Rodriguez et al., 2004). Oscillatory behavior in the visual cortex may correlate strongly with higher order functions such as attentional processing (Muller et al., 2000; Fries et al., 2001), learning (Miltner et al., 1999; Gruber et al., 2001; Gruber et al., 2002) and memory (Tallon-Baudry et al., 1998), therefore directly implicating ACh in the regulation of these processes.
Was this article helpful?