haut de page






















haut de page

mise à jour du 7 mars 2002
 Acta Neurobiologiae Experimentalis
1991, 51, 97-105
télécharger pdf de cet article
Food anticipatory yawning rhythm in the rat
B Holmgren, R Budelli, R Urba-Holmgren, JR Eguibar, M Holmgren, G Baz-Tellez, J Anias
Universidad de Puebla , Puebla, Mexico
Tous les travaux de MR Melis & A Argiolas 
Tous les travaux de M Eguibar & G Holmgren
Introduction : A brief communication in this same journal (Anias et al. 1984) described the existence of circadian variations in the frequency of spontaneous yawning in the rat. The observations were made on a line of Sprague-Dawley rats, genetically selected to obtain a higher incidence of yawning (HY rats) (Urba-Holmgren et al. 1990). In that report an effort was made to correlate diurnal changes in yawning frequency with circadian variations in activity (as reviewed in the literature) of the neurotransmitter systems that had been postulated as participating in the central control and regulation of yawning (Anias et al. 1984). Since yawning frequency was highest during the last light hour, both under natural and artificial light-dark (LD) illumination schedules, it was suggested that the LD transition might be the "primary synchronizer" of the circadian rhythm of yawning. But that suggestion was not supported by a formal differentiation between circadian rhythms proper, which are capable of free-running in absence of the entraining stimuli, and daily or diurnal rhythms, which "are induced and dependent upon rhythmic occurrence of environmental events" (Moore 1980). Nor had the possibility been explored that yawning behaviour might be entrainable to "secondary synchronizing" stimuli, as restricted food or water availability which had been proved to be important synchronizers of several behavioural rhythms (Krieger 1974, Edmonds and Adler 1977, Krieger et al. 1977, Morimoto et al. 1977, Takahashi et al. 1977, Krieger and Hauser 1978, Sulzman et al. 1978, Morimoto et al. 1979, Phillips and Mikulka 1979, Boulos 1980, Moore 1980, Coleman et al. 1982, Dhume and Cogate 1982, Inouye 1982, Honma et al. 1983, Mistlberger and Rechtschaffen 1984) (for recent reviews see Rusak 1981, Takahashi and Katz 1982, Hiroshige 1984, Turek 1985, Rosenwasser and Adler 1986). The purpose of the present work is to study spontaneous yawning behaviour in rats maintained under continuous illumination, and examine if food availability, regularly restricted to 2 or 3 daytime hours can entrain a rhythmic variation in yawning, both in rats under constant light, and in animals under normal 12-12 LD cycles.

Methods : This study is based on the observation of several groups of male Sprague-Dawley rats (46 animals in total), from the 8th, 9th, 13th and 16th generations of the HY subline, genetically selected at our Animal House. Their general housing and feeding conditions and the method we use for the observation of yawning have already been described (Anias et al. 1984, Holmgren et al. 1985). At the preliminary stage we observed yawning for 24 h a day in sessions of différent durations irregularly distributed throughout several days. But as this procedure seemed to introduce unwanted variability, in each experimental condition yawning was at the end monitored during an uninterrupted session of 24-26 h. Finally, because of the inadequateness of averaging data from différent animals in a free running condition, we introduced a balanced experimental design in which each, animal was its own control and we monitored-yawning continuously during 56-60 h on two occasions,separated by 22 days. Different group of animals were studied exclusively under one of the following conditions (A-D):

A. Constant light and ad libitum feeding The observation of yawning behaviour, under constant illumination (LL), was done in twelve animals (two rats per cage) from the 8th and 9th generations of the HY subline. At the age of 2 months they were transferred from the Animal House to a laboratory observation room of 18 m' permanently illuminated with two 80 watts 2.20 m long fluorescent tubes from a 4 m high celling. The room was not exclusive for the experimental animals: during the day it was randomly used for observation of other animals and study purposes. Although silence was recommended, slight noise and personnel traff ic were inevitable between 08:00 and 19:00. After 3 weeks of adaptation to the LL regime, with food and water available ad libitum, yawning occurrence was monitored continuously during 24 h by trained couples of observers replacing each other every 2 h. Finally, two additional groups of four rats each, from HY F13, after the same period of adaptation to LL, were subject to yawning monitoring in uninterrupted observation sessions of 24 h.

B. Restricted daytime feeding in animalsunder constant illumination For this purpose the rats were changerd to clean boxes every day, at regular intervals before giving them their daily ration of standard laboratory rodent pellets (Purina), placed directly on the layer of wood shavings covering the floor of the boxes. During the adaptation period to restricted feeding, which lasted 18 days, the animals were sometimes weighed before and after eating, in order to estimate their daily food intake. Excess of food was rernoved from the cages at the end of the feeding time. The animals had free access to water over 24 h. Two groups of 4 rats each and one of 6 animals were subject to three slightly différent feeding and observation schedules: one group had food avallable during 3 h (14.00 to 17.00) and the observation sessions took place over 30 days after the initial period of adaptation to this regime; the second group had a restricted feeding time (RFT) of only two hours (12.00 to 14.00),the observations being made during 45 days after the adaptation period; the third group had food available during two and a half hours (08:00 to 10:30) and was observed continuously in a single session of 26 h, which began with the animals eating their daily ration of food pellets, and ended with the rats fasting until 10:00 next day.

C. Restricted feeding ame in rats under a LD schedule A group of six two month-old HY rats, from the 9th generation, housed in two cages with three animals per cage, were adapted to a LD cycle (lights on from 7:00 to 19:00) with food available between 9:00 and 11:00. Once adapted to this regime for 3 weeks, the rats were continuously observed and yawning monitored during 26 h, starting from 9:00 one day to 11:00 next day.

D. Balanced experimental design with animais being their own controls In these experiments, we used six two month-old HY-males from generation F16, randomly distributed in two groups of three animals each and treated under continuous illumination according to the following feeding schedule.

E. Fasting experiments After completing the observation of yawning along 24 h of theday, some of the animal groups fasted during 48 to 96 h, with yawning monitored several times during that period in order to study the evolution of spontaneous and drug induced yawning under fasting conditions. These experiments will be described in full separately.



The initial purpose of this work was to establish whether the daily rhythm in yawning behaviour described in the rat, with its peak in frequency preceding the LD transition (Amas et al. 1984) could bc properly considered as circadian, i.e. as an endogenous rhythm governed by a circadian pacemaker, entrainable to the LD transition and capable of free-rurming in its absence. In our experiments with rats adapted to live under constant light, and fed ad libitum, yawning activity exhibits several peaks during the 24 h, as if under aperiodic environmental or internal underlying influences, rather than under the regular control of a circadian pacemaker. The vectors representing maximal yawning hour intervals of the différent animals over 24 h seem to be randomly distributed.

The loss in the diurnal rhythm of yawning behaviour in rats exposed to constant illumination while fed ad libitum does not necessarily mean that this rhythms is not endogenous. Several well documented circadian rhythm, as those of plasma corticosteroid concentration, body temperature levels and food consurription, which in rats are entrainable to the LD cycle, also tend to disappear under constant dim light (Takahashi et al. 1977, Honma and Hiroshige 1978a, 1978b, Morimoto et al. 1979) prolonged for more than one month. The importance of the time of feeding in modifying behavioural rhythms has been known for a long time. Honma et al. (1983) recalled that already in 1922 Richter described an increase in locomotor activity in rats before the time of feeding. Restricted periodical feeding has more recently been used by many authors and in different animals to entrain other biological rhythms: locomotion, running activity, body temperature, drinking and urinary excretion, adrenocortical activity, etc..(Krieger 1974, Edmonds and Adler 1977, Krieger et al. 1977, Morimoto et al. 1977, Takahashi et al. 1977, Phillips and Mikulka 1979, Boulos and Terman 1980, Moore 1980, Coleman et al. 1982, Inouye 1982, Honma et al. 1983, Hiroshige 1984, Mistlberger and Rechtschaffen 1984, Shiraishi et al. 1984). Our expeYiments with HY rats under constant light and only one regular daily meal, show that after 3 weeks' exposure to this regime, a significant peak in yawning activity anticipates the time of food availability.

In captive African lions and mandrils in a Zoo, subjected to regular feeding times, Baenninger (1987) has recently described a clear peak in yawning during the last hour before the animals recelved their food. But this does not necqsarily mean that yawning might be a sign of hunger, as suggested by Barbizet (1958). If fasting was prolonged for more than 24 h spontaneous yawning diminished, and practically disappeared if the animals fasted 3 or 4 days. Under these last conditions even apomorphine-induced yawning was completely blocked (Anias and Holmgren, unpublished results). As to the relative potencies of the L-D transition and the restricted meal time as entrainers of the yawning daily rhythm, the experiment illustrated in Fig. 2C seems quite convincing: the peak in yawning just before dark is absent. These results confirin the opinions of other authors that the time of food presentation (Edmonds and Adler 1977, Krieger and Hauser 1978, Sulzman et al. 1978) or of water availability (Dhume and Cogate 1982) may be more potent synchronizers than the L-D cycle, in relation to several rhythmic physiological or behavioural variables. Since rats under a LD and free feeding schedule eat mostly during the early dark hours, this eating habit might be permanently reinforcing the tendency for a pre-dark peak in yawning to appear, making the L-D transition look like a stronger synchronizer than what it really is.

Yawning has commonly been considered to be related to drowsiness preceding or following sleep (Barbizet 1958). But increased yawning behaviour, anticipatory to the restricted feeding time in rats, seems to coincide with a state of alertness of the animals, with increased locomotor activity (Phillips and Mikulka 1979) or wheel running (Edmonds and Adler 1977, Coleman et al. 1982, Honma et al. 1983) and higher plasma levels of corticosteroids (Krieger 1974, Morimoto et al. 1977). Food "expectancy" might, in a certain way, be considered as a sort of psycho] ogical stress or situational anxiety, condition which recently, in a very brief editorial comment (Shader and Greenblatt 1985), has been mentioned to be positively correlated with yawning in humans.