Akiyoshi et al. [1] reported that canine plasma CgA immunologically cross-reacts with human CgA and that measurement of the canine plasma CgA concentration may provide a useful index for evaluating the acute stress response. However, 24-h internal secretion patterns must be considered since the circadian rhythm might influence CgA concentrations. To our knowledge, there has been no report concerning the circadian rhythms of plasma and salivary CgA concentrations in dogs. The advantage of using saliva is that it is obtainable without the discomfort of blood sampling. If there is a circadian rhythm in canine salivary CgA secretion, then such a pattern should be considered in future research on dog stress. The purpose of this investigation was to determine if salivary CgA secretion in dogs has a circadian rhythm.

Sixteen healthy beagle dogs (8 male and 8 female; age: 3.40 ± 0.59 years) were used. Dogs were housed in the sampling room for three days for adaptation. The lights were turned on from 8:00 a.m. to 8:00 p.m., and the room temperature was kept between 20 and 25℃ throughout the experiment. Water was given ad libitum, and dry food was provided at 8:00 a.m. The dry food was provided after saliva sampling. This experiment was approved by the Ethics Committee on Animal Experiments at the School of Veterinary Medicine and Animal Sciences, Kitasato University.

Experimental saliva sampling was performed during three nonconsecutive 24-h periods. The sixteen dogs were moved to the sampling room and housed individually in cages. Salivary sampling was done at 4-h intervals from 12:00 a.m. and 12:00 p.m. of the following day. However, in order to habituate dogs to saliva sampling, we sampled their saliva hourly starting 3 h before the experiment began. At night, saliva sampling was performed using a dim light.

Saliva was collected by inserting a cotton ball tied with a cotton thread into the dog's oral cavity for 1 min. The cotton ball was then suspended by a thread in a 10-ml centrifuge tube so it did not reach the bottom and was immediately centrifuged at 3,000 g at 4℃ for 15 min. After centrifugation, saliva samples were stored at -20℃ until analysis was performed. Salivary CgA concentrations were measured using a Human CgA enzyme-linked immunosorbent assay kit (Yanaihara Institute, Japan). All samples were analyzed in duplicate. Salivary CgA concentrations were measured as picomoles of CgA per milligram of protein.

Values are expressed as means ± SD. Repeated measures ANOVA and Scheffe's test were used for analysis. Student's unpaired test was conducted to determine the mean values for male and female dogs at each collection time. p values < 0.05 were considered statistically significant.

No circadian rhythm for salivary CgA was observed among the 16 dogs (Fig. 1). The highest concentration (3.28 ± 0.22 pmol/mg) of salivary CgA was observed at 8:00 a.m., and the lowest concentration (3.05 ± 0.28 pmol/mg) was observed at 0:00 a.m. However, no differences were noted in the salivary CgA levels measured every 4 h. There were no differences in salivary CgA concentrations between male and female dogs (Fig. 2). The highest and lowest concentrations of CgA measured in males (3.25 ± 0.20 and 3.00 ± 0.22 pmol/mg) and in females (3.28 ± 0.22 and 3.00 ± 0.28 pmol/mg) were recorded at 8:00 a.m. and 0:00 a.m.

Salivary sampling was performed 3 h before this experiment was started, because the salivary CgA concentration obtained during the first sampling was remarkably high. Moreover, the results were not influenced at the time of salivary sampling. When interpreting these results, we considered the possibility that a dog would undergo a stress reaction when the cotton ball was inserted into its oral cavity. As a result, the salivary CgA level peaked during the first sampling and then decreased 1 h later. Thereafter, the salivary CgA concentration remained low up through 3 h (data not shown). In this study, the dogs exhibited no circadian rhythm in salivary CgA secretion.

To our knowledge, there has been no report in the literature concerning the circadian rhythm of plasma and salivary CgA concentrations in dogs. In humans, plasma CgA secretion has been clearly shown to exhibit no circadian rhythm [4,16]. Recently, Den et al. [4] reported that the human salivary CgA concentration peaked upon awakening, quickly decreased to a nadir after 1 h, then remained at a low level throughout the day. In the present experiment, the highest concentration of salivary CgA was observed at 8:00 a.m., and the lowest level was observed at 0:00 a.m. the following day. This is in contrast to the lack of a circadian rhythm seen with human salivary CgA concentrations. Salivary cortisol concentrations, which serve as a useful stress index, are high in the morning and low in the afternoon in many mammals. However, Koyama et al. [11] reported no circadian rhythm in salivary cortisol concentrations in dogs and proposed that the circadian mechanism in dogs may be basically different from that seen in other species'. Moreover, a number of investigators have failed to detect a circadian rhythm in canine hormones [5-7,10,15,17].

Our study is the first known investigation of the salivary CgA circadian rhythm in dogs. In humans, salivary CgA secretion might be a sensitive index of psychosomatic stress [12]. Akiyoshi et al. [1] indicated that measurement of plasma CgA concentrations might provide a useful index for evaluating the acute stress response in dogs. Therefore, an understanding of the circadian rhythm associated with salivary CgA secretion in normal dogs is also important. Additional studies are warranted to better understand salivary CgA concentrations and their relationship to the canine stress response.