Motion sickness is a common disorder induced by provocative motion of transports and simulators or by weightlessness and virtual reality (Golding and Gresty, 2015). Although several hypotheses have been proposed to explain motion sickness, the precise biological mechanism of the individual difference in motion sickness susceptibility has not been clarified yet (Zhang et al., 2016). Accumulating evidence indicates that sex and age are two main predictors of motion sickness susceptibility in general populations (Golding, 2006). Women always report greater incidence in motion sickness history, (Flanagan et al., 2005) and seem to be more susceptible to seasickness, carsickness, simulator sickness, and visually induced motion sickness than men (Lawther and Griffin, 1988; Turner and Griffin, 1999; Flanagan et al., 2005). However, the hypotheses of motion sickness etiology such as sensory conflict theory cannot explain gender differences in motion sickness. Evidence-based clinical investigations showed that female gender was also a strong risk factor for chemotherapy-induced and postoperative nausea and vomiting (Golding, 2006; Jordan et al., 2014). These observations seem to suggest that there must be anatomical and/or physiological basis underlying high motion sickness susceptibility in females. However, women are more inclined to admit subjective symptoms (such as dizziness, headache, and nausea) and remember past motion sickness experience (Cheung and Hofer, 2002), thus the social and/or psychological factors cannot be ruled out. In addition, motion sickness susceptibility also fluctuates across age. In general, humans around 6 to 7 years of age begin to feel motion sick during traveling (Reason, 1969). The susceptibility peaks around 9 to 10 years and subsequently declines from the teenage years toward adulthood (Turner and Griffin, 1999). However, whether the age-related decline in motion sickness susceptibility can be attributed to habituation induced by repeated motion exposure during aging is still unclear.

Although motion sickness susceptibility-related factors have been widely and frequently studied in humans, the sex and age differences in motion sickness-related behaviors in rodents have not been fully examined yet. In humans, motion sickness is characterized by subjective symptoms such as drowsiness, headache, stomach discomfort, and nausea as well as objective signs such as pallor, thermal effects and vomiting. Retching and vomiting, as the most pronounced symptom of motion sickness, have always been used to assess motion sickness in Suncus murinus, cats, dogs, and humans (Conder et al., 2008; Percie du Sert et al., 2010; Lackner, 2014; Lucot, 2016). Although laboratory rodents lack the ability of vomit, the majority of rats are prone to display pica (McCaffrey, 1985; Horn et al., 2013), defecation response (Ossenkopp and Frisken, 1982), as well as spontaneous locomotion reduction during or after motion sickness (McCaffrey, 1985; Ossenkopp et al., 1994). However, pica seems to be a self-medicative behavior rather than a sensitive indicator of motion sickness due to its delay in response to motion stimulation and prolonged recovery after motion sickness habituation in rodents (Li et al., 2008; De Jonghe et al., 2009; Wang et al., 2012). Our previous study showed that scopolamine treatment and bilateral labyrinthectomy completely abolished rotation-induced defecation incontinence and hypoactivity, indicating the reliability of these two motion sickness assays (Wang et al., 2015). Recent studies revealed that the emetic drugs- or motion stimulation-paired context, taste and/or odor can produce a characteristic gaping reaction in rats (Parker and Limebeer, 2006; Limebeer et al., 2008). The finding that labyrinthectomy eliminated body rotation-induced but not lithium-induced conditioned rejection responses supports the validity of a rat model of motion sickness (Cordick et al., 1999; Ossenkopp et al., 2003). In addition to behavioral reactions, motion sickness also induced temporal changes in concentration of blood hormones including adrenocorticotropic hormone (ACTH), corticosterone, norepinephrine, epinephrine, β-endorphin and arginine-vasopressin (AVP) in rodents (Otto et al., 2006; Mo et al., 2012). Furthermore, Fos protein, as an indicator of neuronal activity, might also be a molecular marker for MS development and habituation (Abe et al., 2010; Cai et al., 2010). Studies conducted in a variety of animal species found that provocative motion stimulation enhanced Fos protein expression in vestibulo-autonomic nuclei including the caudal medial and spinal vestibular nucleus (MVN and SpVN), the nucleus of solitary tract (NTS), the parabrachial nucleus (PBN), in the limbic system such as the central and medial nucleus of amygdala (CeA and MeA) and in the areas mediating vestibular stress responses such as the locus ceruleus (LC) and the paraventricular hypothalamus nucleus (PVN) (Nakagawa et al., 2003; Cai et al., 2010; Balaban et al., 2014). Therefore, we hypothesized that the patterns of blood hormone responses and central Fos expression might contribute to sex and age differences in motion sickness.

In this study, we tried to investigate whether there are sex and age differences in motion sickness in rats exposed to Ferris wheel-like rotation. Motion sickness-related gastrointestinal responses were assessed by measuring conditioned gaping and defecation responses; whereas the motor and balance impairments were evaluated using open field test and balance beam test in adolescent, young-adult and middle-aged male and female animals. In order to explore the endocrine and/or neuronal activity correlations of sex and age differences in motion sickness responses, we also examined the blood stress hormone levels and observed Fos protein expression pattern after rotation in male and female rats across ages.

The present study revealed that both LiCl injection and rotation stimulation significantly induced conditioned gaping (nausea-like behavior) in both male and female animals, but no significant sex differences were found when females were not in menstruation phase. These results are consistent with several laboratory observations in humans which showed no gender difference in susceptibility of circular vection drum or Coriolis cross-coupling stimulation-induced motion sickness indicated by nausea ratings (Cheung and Hofer, 2002; Klosterhalfen et al., 2006). In a study of anurans, parabolic flight significantly induced emesis in the tree frog R. schlegelii, but no sex differences was also observed (Naitoh et al., 2001). Nevertheless, studies on motion-induced emesis in adult Suncus murinus found more frequent emesis in male animals exposed to 5-min reciprocal shaking, while female animals were more responsive to a much longer stimulation (10 min) with similar amplitude and frequency (Matsuki et al., 1997; Javid and Naylor, 1999). These results clearly demonstrated that females do not always exhibit motion sickness hypersensitivity in either humans or animals. Golding et al. (2005) have reported that motion sickness susceptibility fluctuates through the menstrual cycle in women, and the magnitude of fluctuation in susceptibility across the cycle contributes to around one-third of the overall difference between male and female. Estrous cycle may also influence spontaneous locomotor activity measurements in female rats (Martin and Battig, 1980). The limitation of the present study is that we did not track the estrous cycle and distribution of the estrous cycle was not identified within each group on conditioning and test days. Recently, Cloutier et al. (2017) have confirmed the validity of using conditioned gaping response as an index of nausea in rats. In this study, the conditioning episodes were carried out during the light cycle every 3 days which ensured a different estrous cycle day for females on each of the four conditioning days, and animals in each group were pseudo-randomized across the estrous cycles. They found that female rats had more LiCl-induced gapings and other aversive disgust-related behaviors (forelimb flailing, chin rubbing, and paw treading) than the males. Based on the fact that motion sickness susceptibility might increase during menstruation, it is plausible to presume that female animals within sickness sensitive estrous cycle phase may contribute to high LiCl-induced gaping reactions. Recently, motion sickness-induced hypothermia has been reported in rodents and musk shrews, indicating that thermoregulatory symptoms may serve as physiological correlates of nausea (Nobel et al., 2012; Del Vecchio et al., 2014; Ngampramuan et al., 2014). Gastric myoelectric activity is also associated with nausea in ferrets, Suncus murinus, and humans (Kiernan et al., 1997; Percie du Sert et al., 2010; Lu et al., 2014). These physiological variables should be assessed as objective measurements of nausea in the future. Furthermore, previous studies showed that overshadowing by salient stimuli and repetitive context pre-exposure (latent inhibition) can reduce anticipatory nausea in humans and conditioned aversion in rodents, suggesting that there are psychological effects influencing anticipated nausea and vomiting (Klosterhalfen et al., 2005; Hall and Symonds, 2006; Stockhorst et al., 2014). In the meantime, susceptibility to motion sickness was also found to be associated with certain personality characteristics such as trait-anxiety and neuroticism which appeared to be correlated significantly with high motion sickness susceptibility in women (Collins and Lentz, 1977; Bick, 1983; Paillard et al., 2013). Furthermore, manipulation of expectation affected not only motion-induced nausea symptoms but also gastric physiological responses, suggesting an unspecified psychophysiological interaction mechanism responsible for the susceptibility of nausea and emesis to motion stimulation and other nauseogenic stimuli (Williamson et al., 2004; Levine et al., 2006). It is also noteworthy that expectancy for sickness was more effective to enhance tolerance to rotation chair stimulation in men than women (Klosterhalfen et al., 2005; Klosterhalfen et al., 2009). Although women report more overall symptoms of visually induced motion sickness and significantly more gastrointestinal symptoms than males, gastric tachyarrhythmia, as a biological markers of nausea, did not show sex differences accordingly (Jokerst et al., 1999). These evidence suggested that the sex differences in motion sickness may not only arise from menstrual/hormonal cycles but also can be influenced by psychological factors and might reflects differences in subjective awareness of the illness to motion and other nauseogenic stimuli.(Golding et al., 2005).

Anatomical evidence supports the view that vestibular stress such as motion sickness may modulate hypothalamic-pituitary-adrenal (HPA) axis through the vestibulo-paraventricular pathways (Markia et al., 2008). The present study also found that female animals had significantly higher defecation responses after rotation than males did. It seems that females tend to show heightened stress responses than males after rotation (Curtis et al., 2006; Goel and Bale, 2009). However, we revealed that rotation only significantly affected blood ACTH and corticosterone levels in male young adults, and significant gender difference on blood hormone responses after rotation was observed across ages. Although, we did not observe hormone and defecation responses as a function of time during rotation, there is sufficient evidence supporting that temporal changes of blood stress hormones did not synchronize those of motion-induced nausea, and that the activation of HPA axis might be a transient response to provocative motion rather than a causing factor of motion-induced autonomic responses (Otto et al., 2006). In addition, it is also noteworthy that female animals always exhibited greater behavioral responses to stress, such as increased immobile time in the forced swim and tail suspension tests, compared to males (Bale and Vale, 2003; Altemus, 2006). Similarly, the present study also found that there were great sex differences in spontaneous locomotion activity in middle-aged animals. However, motion sickness can induce neurosensory, sensory-motor and perceptual disturbances which might also impare motor performance immediately after the cessation of motion stimulation (Reschke et al., 1998). Recent studies have confirmed that spatial orientation preference had a significant relationship with susceptibility to simulated motion sickness symptoms (Chen et al., 2015, 2016). These evidence suggest that sex differences in motion sickness induced locomotion disorder might also be associated with the diversity in spatial cognition and spatial orientation (Bettis and Jacobs, 2009; Locklear and Kritzer, 2014). Furthermore, balance performance in females was not different from males during training period in the present study. Male and female animals exhibited almost equivalent perturbations in motor coordination after rotation during adolescent and young adulthood. Although there are great differences in the spatial magnitude of postural sway and the control of posture in humans (Koslucher et al., 2016a,b), our results suggest that quadrupeds (e.g., rodents) seem to be more stable than bipeds (e.g., humans) in both anteroposterior and mediolateral body axes and postural sway might not be a common factor determining motion sickness susceptibility across species (Takahashi et al., 1991).

The present study also found that rotation-induced gastrointestinal responses (conditioned gaping and defecation responses) declined during aging in both males and females. These results are consistent with previous studies which showed that rotation-induced pica decreased with age in rodents (McCaffrey and Graham, 1980), and middle-aged and aged subjects were more resistant to motion sickness than adolescents and young adults (Cooper et al., 1997; Gahlinger, 2000). Since animals had never received motion challenges before experiment, it clearly suggested that the age-related differences are mainly attributed to physiological aging process but not habituation that was commonly seen in humans. In the meantime, the age-related changes in blood levels of ACTH, corticosterone, norepinephrine and epinephrine are also consistent with previous findings (Hauger et al., 1994; Ivanisevic-Milovanovic et al., 1997). However, these changes might not contribute to age differences in motion sickness due to a lack of age × rotation interaction effects. In addition, many studies have found an age-related degeneration of the peripheral vestibular receptor hair cells, the Scarpa’s ganglion neurons and vestibular nucleus neurons in humans and some animal species (Brosel et al., 2016; Smith, 2016). Some studies also found structural changes in cytoplasm such as nuclear membrane invaginations, disorganized endoplasmic reticulum and lipofuscin-like dense bodies in vestibular nucleus neurons of rodents (Takeuchi et al., 1997; Fernandez et al., 2007). However, our study found that rotation-induced Fos expression was significantly increased in the SpVN, the NTS and the PBN as well as in the CeA and the MeA in an age-dependant manner, indicating that neuronal activity in these vestibular and autonomic nuclei might not be influenced by the pathological changes during aging. Our observations also coincide with previous findings regarding age-related differences in central Fos expression. For example, adult mice treated with lipopolysaccharide (LPS) display significantly greater Fos immunoreactivity compared to their saline controls while adolescent mice show no difference in Fos expression between LPS- and saline-treated mice (Girard-Joyal et al., 2015). Pubertal rats also showed reduced Fos expression in several brain regions compared to adult animals following a 2 h restraint stress (Kellogg et al., 1998). Meanwhile, it is widely accepted that the caudal aspect of vestibular nuclei including the MVN and the SpVN participates in generating motion-induced autonomic responses in rodents (Yates et al., 2014). Our previous study showed that direct pathways from the MVN and the SpVN to the NTS and the PBN were specifically activated by rotation stimulation in adult rats (Cai et al., 2007). Recent studies have demonstrated that the PBN was necessary for the anorexigenic effects of LiCl and for CTA acquisition in mice and the NTS-PBN-CeA connections mediated cisplatin-induced pica, anorexia and loss of body weight in rats (Alhadeff et al., 2015; Carter et al., 2015; Roman et al., 2016). Meanwhile, cisplatin injection or hypergravity stimulation also increased Fos expression in rat CeA and MeA and in the ferrets receiving Exendin-4 injection (Nakagawa et al., 2003; Horn et al., 2007; Lu et al., 2016). Functional magnetic resonance imaging studies found that the amygdala was activated in humans experiencing motion sickness-induced nausea or mal de debarquement syndrome (Cha et al., 2012; Napadow et al., 2013). Moreover, our study also found that the middle-aged animals had much greater decline in the rotation-induced conditioned gaping reactions than the LiCl-induced ones. Middle-aged males with greater reduction in locomotion responses after rotation also had lower numbers of Fos-LI neurons in the NTS, the PBN, as well as in the LC and the PVN than female animals. These evidences strongly support the viewpoint that increased Fos expression in these vestibular and autonomic nuclei might correlate with the decreased rotation-related behavioral responses during aging in the rats. However, whether these activated neurons are involved in regulation of intrinsic “homeostasis” after motion stimulation is still unclear and needs further investigation (Martinez et al., 2002).

Our previous study found that rotation stimulation significantly reduced both horizontal and vertical spontaneous motor activity in rats, but the change of horizontal movement seemed to be more pronounced than the vertical movement (Wang et al., 2012). In contrast, rats with chemically induced vestibular dysfunction exhibited more open-field ambulation but fewer rearing responses compared with controls (Ossenkopp et al., 1990, 1992). The current spontaneous locomotion observation also coincides with previous results showing hypoactivity in rodent motion sickness model (Ossenkopp et al., 1994; Santucci et al., 2009). In addition, the present study revealed an age-related alleviation in rotation-induced motor disorders (i.e., hypoactivity and balance disturbance) in the rats. The results of balance beam test are inconsistent with a previous observation showing that the impact of rotation stimulation (without inducing overt signs of motion sickness) on balance beam performance is significantly greater in 9-month-old and 13-month-old mice than in 1-month-old mice (Tung et al., 2014). The discrepancy can be explained mainly by several factors such as species, rotation pattern and training procedure. As we observed similar age-related alterations in males and females, it is reasonable to conclude that the impact of age on vestibular stimulation-induced motor disorders might possibly depend on whether or not the motion sickness has been triggered. Furthermore, our results also suggested that the SpVN might play a role in age dependent motor disorders-induced by motion sickness. This notion can be supported by the fact that medial vestibulospinal tracts in part originates from the SpVN in rodents (Liang et al., 2015). Immunohistochemical studies confirmed the existence of an inhibitory influence of SpVN to the spinal cord (Blessing et al., 1987; Valla et al., 2003). Further investigations should be performed to identify the precise neuron population in the SpVN that is involved in age-related regulation of autonomic and/or locomotor responses during motion sickness.

In summary, we found an age-related alleviation in motion sickness-related gastrointestinal responses (nausea and defecation) and motor disorders (hypoactivity and balance disturbance) especially in middle-aged rats. Females only had higher defecation responses during adolescent and young-adult period than males. Although there were significant age-related alterations in several blood hormones (ACTH, corticosterone, norepinephrine, and epinephrine), such effects are irrelevant to rotation in both males and females. The present study also found remarkable age differences in Fos expression pattern in nausea-related brain structure including the SpVN, the NTS, the PBN, and the amygdala, with a significant increase in Fos levels in the middle-aged animals compared with the adolescents and the young-adults. Females showed higher Fos levels in the NTS, the PBN, the LC, and the PVN during young-adult and/or middle-aged period than males. These findings provide the evidence that the sex and age differences in provocative motion-induced gastrointestinal responses and motor disorders may not correlate with stress hormone responses and habituation. The age-related decline in motion sickness susceptibility might be mainly attributed to the activity changes in neurons of the vestibular and autonomic areas including the SpVN, the NTS, the PBN, and the amygdala.

WZ, JW, LP, and RQ performed behavioral experiments. WZ, JW, and LP were responsible for the stress hormone measurements. LP, WZ, and JW completed immunohistochemistry experiment. JL and PL were responsible for animal breeding. WZ, JW, and YC were responsible for the design of the work. LP, PL, and WZ were responsible for data analysis. YC, WZ, and JW were responsible for writing the manuscript. All authors read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.