Eyelid position and the electromyographic activity of the orbicularis oculi muscle were recorded unilaterally in rabbits during reflex and conditioned blinks. Air-puff-evoked blinks consisted of a fast downward phase followed sometimes by successive downward sags. The reopening phase had a much longer duration and slower peak velocity. Onset latency, maximum amplitude, peak velocity, and rise time of reflex blinks depended on the intensity and duration of the air puff-evoking stimulus. A flashlight focused on the eye also evoked reflex blinks, but not flashes of light, or tones. Both delayed and trace classical conditioning paradigms were used. For delayed conditioning, animals were presented with a 350-ms, 90-dB, 600-Hz tone, as conditioned stimulus (CS). For trace conditioning, animals were presented with a 10-ms, 1-k/cm(2) air puff, as CS. The unconditioned stimulus (US) consisted of a 100-ms, 3-k/cm(2) air puff. The stimulus interval between CS and US onsets was 250 ms. Conditioned responses (CRs) to tones were composed of downward sags that increased in number through the successive conditioning sessions. The onset latency of the CR decreased across conditioning at the same time as its maximum amplitude and its peak velocity increased, but the time-to-peak of the CR remained unaltered. The topography of CRs evoked by short, weak air puffs as the CS showed three different components: the alpha response to the CS, the CR, and the reflex response to the US. Through conditioning, CRs showed a decrease in onset latency, and an increase in maximum amplitude and peak velocity. The time-to-peak of the CR remained unchanged. A power spectrum analysis of reflex and conditioned blink acceleration profiles showed a significant approximately 8-Hz oscillation within a broadband of frequencies between 4 and 15 Hz. Nose and mandible movements presented power spectrum profiles different from those characterizing reflex and conditioned blinks. It is concluded that eyelid reflex responses in the rabbit present significant differences from CRs in their profiles and metric properties, suggesting different neural origins, but that a common approximately 8-Hz neural oscillator underlies lid motor performance. According to available data, the frequency of this putative oscillator seems to be related to the species size.

Neuronal nicotinic acetylcholine receptors from bovine adrenomedullary chromaffin cells play a primary role in triggering catecholamine secretion. In the present study, their constituent subunits were characterized. In addition to the cx3 subunit, which we have previously cloned, the presence of a5 and /94 but not of /32 subunits was detected by reverse transcription-PCR analysis of mRNA from adrenal medulla. In situ hybridization indicated that cr3, cr5, and~34subunits are coexpressed in all chromaffin cells. The primary structure of a5 and /34 subunits was determined and functional receptors were obtained upon coinjection of subunit cRNAs into Xenopus oocytes. In contrast to other /34-containing nicotinic receptors, the ones formed by the bovine /94 subunit are insensitive to the agonist cytisine. Finally, we characterized the intergenic region of cr3 and cr5 subunits, which together with the /34 subunit, form a gene cluster in rats and chickens. RNase assays and the existence of overlapping cDNAs indicate that, in the bovine genome, the cr3 and cr5 genes overlap at their 3' ends. This fact is probably due to inefficient transcription termination, as a result of weak polyadenylation signals.

Adrenomedullary chromaffin cells express at least two subtypes of acetylcholine nicotinic receptors, which differ in their sensitivity to the snake toxin ␣-bungarotoxin. One subtype is involved in the activation step of the catecholamine secretion process and is not blocked by the toxin. The other is ␣-bungarotoxin-sensitive, and its functional role has not yet been defined. The ␣7 subunit is a component of this subtype. Autoradiography of bovine adrenal gland slices with ␣-bungarotoxin indicates that these receptors are restricted to medullary areas adjacent to the adrenal cortex and colocalize with the enzyme phenylethanolamine N-methyl transferase (PNMT), which confers the adrenergic phenotype to chromaffin cells. Transcripts corresponding to the ␣7 subunit also are localized exclusively to adrenergic cells. To identify possible transcriptional regulatory elements of the ␣7 subunit gene involved in the restricted expression of nicotinic receptors, we isolated and characterized its 5Ј flanking region, revealing putative binding sites for the immediate early gene transcription factor Egr-1, which is known to activate PNMT expression. In reporter gene transfection experiments, Egr-1 increased ␣7 promoter activity by up to sevenfold. Activation was abolished when the most promoter-proximal of the Egr-1 sites was mutated, whereas modification of a close upstream site produced a partial decrease of the Egr-1 response. Because Egr-1 was found to be expressed exclusively in adrenergic cells, we suggest that this transcription factor may be part of a common mechanism involved in the induction of the adrenergic phenotype and the differential expression of ␣-bungarotoxin-sensitive nicotinic receptors in the adrenal gland.

During embryonic development, restricted expression of the regulatory genes Krox20 and kreisler are involved in segmentation and antero‐posterior patterning of the hindbrain neural tube. The analysis of transgenic mice in which specific rhombomeres (r) are eliminated points to an important role of segmentation in the generation of neuronal networks controlling vital rhythmic behaviours such as respiration. Thus, elimination of r3 and r5 in Krox20–/– mice suppresses a pontine antiapneic system (Jacquin et al., 1996). We now compare Krox20–/– to kreisler heterozygous (+/kr) and homozygous (kr/kr) mutant neonates. In +/kr mutant mice, we describe hyperactivity of the antiapneic system: analysis of rhythm generation in vitro revealed a pontine modification in keeping with abnormal cell specifications previously reported in r3 (Manzanares et al., 1999b). In kr/kr mice, elimination of r5 abolished all +/kr respiratory traits, suggesting that +/kr hyperactivity of the antiapneic system is mediated through r5‐derived territories. Furthermore, collateral chemosensory pathways that normally mediate delayed responses to hypoxia and hyperoxia were not functional in kr/kr mice. We conclude that the pontine antiapneic system originates from r3r4, but not r5. A different rhythm‐promoting system originates in r5 and kreisler controls the development of antiapneic and chemosensory signal transmission at this level.