During the first months of postnatal life kittens undergo a critical period in which neurons in the visual cortex show a remarkable modifiability in response to changing visual stimuli (Hubel and Wiesel, 1970; Olson and Freeman, 1980; Cynader et al., 1980; Cynader and Shaw, 1986). Several reports suggest that this modifiability only occurs in cooperation with attentional mechanisms mediated by monoaminergic and/or cholinergic systems which substantially innervate the visual cortex (Kasamatsu et al., 1979; Bear and Singer, 1986). To learn more about the cellular mechanisms involved in visual cortex plasticity a systematic study of the development of the localisation and characteristics (Bmax and KD) of the different neurotransmitter receptors in the developing cortex, was carried out.

The rules underlying muscarinic acetylcholine receptor (mAChR) regulation in an in vitro cortical slice preparation of adult rats were examined following various alterations of bioelectric activity and following agonist stimulation. Muscarinic ACh antagonists [3H]N-methyl scopolamine ([3H]NMS) or [3H]quinuclidinyl benzylate ([3H]QNB) were used to label cell surface vs total (i.e. surface and internal) receptors, respectively. Depolarization of neural membranes for 4 h at 22-37 degrees C using veratridine or high external potassium (K+) led to a temperature-dependent down-regulation of surface mAChR of 26.2% and 11.3%. Total mAChRs decreased by 37.6% and 8.1%. Addition of picrotoxin and glutamic acid also led to decreases in mAChRs. Increases in inward chloride ion current induced by gamma-aminobutyric acid (GABA) or gold chloride had no significant effect on mAChRs. Blockade of calcium channels and synaptic transmission by magnesium or cobalt and postsynaptic calcium channels with nifedipine showed a significant effect on mAChRs only in the latter case. In contrast, agonist stimulation using carbachol led to a large down-regulation for both [3H]NMS and [3H]QNB (26.1%, 35.9%). ACh decreased [3H]QNB binding by 33.9%, but had little effect on [3H]NMS binding (6.3%). For [3H]QNB binding sites the effects of carbachol appeared to summate with those of veratridine. Down-regulation of [3H]NMS labelled mAChRs by carbachol and veratridine had an estimated half-time of 30 min and 2 h, respectively. Neither the effects of veratridine nor carbachol could be antagonized by tetrodotoxin (TTX), showing that the effects were not due to an increase in sodium ion currents. However, a common thread linking the various agents which induce mAChR down-regulation appears to involve changes in potassium (K+) current. Potassium channel blockers tetraethylammonium chloride (TEA), 4-aminopyridine (4-AP) and apamin had little independent effect on mAChR number, but prevented veratridine-induced down-regulation, presumably through a blockade of K+- and Ca2+-dependent K+-channels. Only TEA and 4-AP diminished carbachol-induced down-regulation suggesting that this effect involves only the non Ca2+-dependent K+-channels. It thus appears that mAChR regulation in the rat cerebral cortex is linked to changes in active K+-channel currents: activation of the K+-channel by depolarization-induced changes in K+ current or by agonist stimulation leading to changes in the selective K+ currents stimulate mAChR down-regulation; blockage of the K+-channels prevents this down-regulation.

The development of spontaneous bioelectric activity (SBA) was studied in dissociated occipital cortex cultures prepared from 19 day old rat fetuses. All cultures, recorded one per diem from 5 to 33 days in vitro (div), showed SBA. Computer analysis of 76 extracellularly recorded single unit spike trains was carried out after selection on the basis of stationarity criteria. Statistically significant developmental trends were found in (i) interspike interval dependencies and (ii) fluctuations in mean firing rate, on the order of a minute or longer. The highly dependent firing patterns, including stereotyped bursting, were present mostly in the 9–12 div group, whereas minute-to-minute fluctuations in the intensity of firing were considerably more pronounced in the oldest group (22–33 div) than in the younger cultures. In addition, firing categories defined on the basis of factor-analysis revealed that such fluctuations were almost exclusively to be found in neurons which fired in a pronounced ‘burst’, rather than a relatively continuous fashion. Only a few mature appearing synaptic structures were observed electron microscopically prior to 12 div, but increased steadily in number thereafter. No cultures prior to 14 div, but all cultures older than this, stained positively for the presence of glutamic acid decarboxylase. An extensive immunoreactive, putative GABAergic, network was present by three weeks in vitro.