The effects of neonatal maternal deprivation and chronic unpredictable stresses on migraine-like behaviors in adult rats

BACKGROUND
Stress is known to cause migraine. This study investigates the effects of neonatal maternal deprivation (MD) and chronic unpredictable stress (CUS) on migraine in rats.


METHODS
Seventy rats were randomly divided into ten groups (five groups of each sex, and seven rats/group). The groups included: untreated intact, nitroglycerin (NTG) only, NTG+MD, NTG+CUS (10 weeks after birth), and NTG+MD+CUS. For the induction of MD, pups were separated from their mothers from postnatal day 2 to day 14. The CUS was conducted by daily exposure to different stressors for 2 weeks. For the induction of migraine after stress, NTG (5 mg/kg/IP) was administered every second day for 9 days. Afterward, NTG-related symptoms, including climbing behavior, facial rubbing, body grooming, freezing behavior, and head-scratching, were recorded for 90 minutes. Statistical differences between the groups were analyzed by one-way and two-way ANOVA followed by the Newman-Keuls test.


RESULTS
Migraine symptoms, including increased head-scratching, facial rubbing, and decreased climbing behavior, were more significant in females than in males. Head scratching and facial rubbing increased in stressed females, but not in males as compared to NTG-treated rats. Body grooming was significantly decreased in MD males compared to the NTG group. The effects of NTG in MD+CUS on the rats did not differ from those in the MD or CUS groups.


CONCLUSIONS
MD and CUS had a sex-related aggravating effect on the development of migraine, while the combination of MD and CUS had no additive migraine-aggravating effect.


Introduction
Migraine is a high-prevalence primary headache disorder that negatively impacts the quality of life of sufferers throughout the world [23]. It is characterized most often by a unilateral pulsating episodic headache. Some patients may experience aura symptoms, including vomiting, nausea, dizziness, and photophobia, approximately 10 to 30 min before a migraine headache begins [33].
Typically, headaches are triggered due to a disorder of orofacial sensory inputs at different relay levels. For example, migraine headaches occur as a result of abnormal activation of trigemino-cervical complex (TCC) neurons. Hyperexcitation of TCC neurons innervating cerebral and meningeal blood vessels has been shown to induce local and vascular changes and trigger a migraine attack through the release of vasoactive peptides like calcitonin gene-related peptide (CGRP) in the trigeminal nerves [14,33].
Migraine is more prevalent in women than in men. Neuroimaging studies in patients suffering from migraine have revealed anatomical and functional gender differences in regional brain activity [26,27]. In addition, puberty-related changes in syntheses and concentrations of sex hormones, mostly estrogen, play an important role in the frequency and severity of migraine headaches [12,36].
There is a complex association between stress levels and the initiation, severity, and frequency of migraine attacks. A number of clinical and preclinical studies suggest that stress is a triggering factor for migraines [22,31]. However, studies have yielded contradictory results. For example, Lipton et al. reported that a decline in stress is associated with migraine onset [24]. Kaufmann and Brennan's study indicated that NTG treatment is not able to induce a significantly more reduction in withdrawal threshold in mice exposed to either social defeat stress or chronic variable stress compared to control animals. The control group and stressed mice did not show any significant difference in the vulnerability to cortical spreading depression or the wave of sustained depolarization that is associated with migraine aura [20].
Maternal deprivation (MD) in the neonatal period of life is associated with profound alterations in neurophysiological responsiveness, including pain perception in adulthood [6,17]. It has been indicated that perinatal stress induces a deleterious effect on the maturation of the sensory-spinal nociceptive system in rats [18].
Neonatal MD might also enhance orofacial mechanical allodynia in adult rats [43]. However, to our knowledge, no data exists on the influence of MD on the pathogenesis of adulthood migraine. Based on the literature, it can be expected that early-life stress poses a risk for migraine attacks in later life. It has been hypothesized that rats that are subjected to chronic unpredictable stress subsequent to MD mightbe more susceptible to migraine. However, the influence of gender on stress and the development of migraine is poorly understood. Taken together, this study hypothesizes that MD stress or later-life chronic unpredictable stress (CUS) is able to trigger or aggravate the NTGinduced migraine-like behavioral indices of adult rats. The combination of different stressors seems to induce more effects on nociceptive behaviors. Finally, we assume that NTG-treated female rats show more behavioral expressions of migraine headaches than male ones.

Animals
Pregnant Wistar rats in the second gestational week were housed in individual cages. The rats were maintained in a fully controlled animal facility (12:12 h light/dark cycle, lights on at 7 am, temperature 22 ± 2 • C). Food and water were available ad libitum. The day of birth was considered postnatal day 0. In total, 70 neonates of both sexes were used in this experiment. The neonates were delivered by twelve mothers with 4-7 pups per labor (mean, 5.6 pups). All the experimental procedures were approved by the Ethical Committee of Kerman University of Medical Sciences (IR.KMU.REC.1397.442).

Experimental group
On postnatal day 2, the pups were randomly divided into ten groups (five groups of each sex and seven rats/group). Both males and females consisted of the following groups: intact group (without treatment), NTG-only group that received no stress paradigm before NTG administration (5 mg/kg/i.p), MD + NTG group that was subjected to MD before NTG administration, CUS + NTG group that was exposed to CUS before NTG administration, and MD + CUS + NTG group that received MD + CUS before NTG administration. In the non-stress condition, pups and their mothers were left undisturbed in the cages for three postnatal weeks. In the MD condition, the pups were separated daily from their dams for 4 h/day (09:30 am to 1:30 pm) up to postnatal day 14 [9]. In MD + CUS conditions, the pups were exposed to a similar maternal absence up to postnatal day 14, followed by later chronic unpredictable stress. All behavioral assessments were done under blind conditions. The experimenters were unaware of the groups and study arrangements. The experimental design is presented in Fig. 1.

Maternal deprivation
For the process of deprivation, the mothers were first removed from their cages and placed in adjacent cages. The pups were then placed in plastic litters provided with nesting material. At the end of the deprivation period, the pups were returned to their respective cages. Once per week, the cages were cleaned, and the bedding was changed. Pups were weaned on postnatal day 15 and housed four to five per cage.

Chronic unpredictable stress
The CUS protocol was designed based on the study by Ma et al., with only minor modifications [1]. Ten-week-old rats were subjected to the following stressors once daily for two weeks: 45 min cage-shaking, 5 min of forced swimming at 10 • C water, 90-min of restraint stress, 24 h water deprivation, 24 h cage-tilting, 24 h food deprivation, or 24 h social isolation (each animal was kept in individual housing). All the rats in each group received these identical stressors during the test period. On each day, the type of stressor was the same for all animals. The stressors were randomly arranged to maintain an unpredictable nature.

Induction and assessment of chronic migraine-associated pain
NTG (5 mg/kg/i.p) was administered every second day (5 injections in total) to induce chronic migraine. The subject rats were taken to the test room one hour before the injection was administered. Immediately after the last injection, migraine-related behaviors were recorded by a video placed in the front of the cage for 90 min. The observed behaviors included: climbing, i.e. front paws seizing the cage; facial rubbing, i.e. movement patterns in which paws contacted facial areas; headscratching in which the rat uses forelimbs to continuously scratch their head; and general body grooming, i.e. movement patterns in which the paws, tongue, and incisors are brought in contact with any part of the body except the face and head [38,42].

Statistical analysis
The data is presented as mean ± standard error of the mean. All the data follows a normal distribution curve. A Statistical analysis of each behavioral test among the stressed or non-stressed groups, in NTG and non-NTG conditions was performed using one-way analysis of variance (ANOVA) followed by Student-Newman-Keuls post hoc tests. In addition, the possible association between sex and the reaction to stress was evaluated by two-way ANOVA. P values smaller than 0.05 were considered significant.

Evaluation of NTG-induced migraine headache symptoms in rats
During the 90-min test period, the numbers of climbing behaviors significantly decreased in both male (p < 0. 01) and female (p < 0. 001) rats treated by NTG as compared to baseline for either sex ( Fig. 2A). NTG-induced migraine significantly increased the head-scratching scores and the time course of facial rubbing responses in male (p < 0. 05) and female rats (p < 0. 001) ( Fig. 2B and 2C). Administering NTG did not change body-grooming behavior in rats [F 3, 27 = 2.141, P = 0.121] (Fig. 2D). Female rats treated by NTG showed an increase in the number of head-scratching and a decrease in observed incidents of climbing behavior compared to male rats (p < 0. 05).
climbing behavior was significantly decreased in both male and female rats exposed to MD or CUS paradigms in comparison to the NTG-only group. However, there was no significant difference in climbing behavior among MD, CUS, or MD + CUS groups for both sexes (Fig. 3A  and 3B).

Evaluation of NTG-induced changes in facial rubbing in rats subjected to stress
NTG-induced facial rubbing time was not significantly different among MD, CUS, and MD + CUS male rats [F 3, 27 = 1.527, P = 0.233] (Fig. 3C). The administration of NTG aggravated facial rubbing behavior in female rats exposed to MD stress or CUS in comparison to groups of rats treated by NTG only or NTG plus MD + CUS (p < 0. 05) (Fig. 3D).

Evaluation of NTG-related head-scratching behavior in rats subjected to stress
As shown in Fig. 4, there are no significant differences in NTGinduced head-scratching behavior in male rats subjected to stress paradigms [F 3, 27 = 0.891, P = 0.460] (Fig. 4A). However, NTG-induced head-scratching was significantly increased in female rats exposed to MD stress or CUS in comparison to the rats treated by NTG (p < 0. 05) (Fig. 4B).

Evaluation of NTG-related grooming behaviors in rats subjected to stress
In male rats exposed to MD or CUS before administration of NTG, the mean time of grooming behavior during the 90 min of the test period was significantly decreased compared to rats only treated with NTG (p < 0.01) (Fig. 5C). Grooming behavior did not significantly change in female migraine sufferers pre-treated with MD, CUS, or combinations of MD and CUS paradigms (Fig. 5D).

Evaluation of migraine-like responses between male and female stress groups
The effect of sex on migraine-like responses was analyzed using a two-way analysis of variance. As shown in Fig. 5A, the administration of NTG significantly decreased cage climbing behavior in CUS-exposed female rats as compared to the male rats [F 5, 42 = 4.220, P = 0.004]. There were significant differences between the different study groups regarding the duration of facial rubbing [F 5, 42 = 9.046, P = 0.0001], body grooming [F 5, 42 = 8.847, P = 0.0001], and the number of headscratching [F 5, 42 = 10.029, P = 0.001]. NTG caused a significant increase in migraine-like responses in female rats exposed to either MD stress or CUS paradigms compared to male rats (p < 0.05) (Fig. 5B-D).

Discussion
This study explores the influence of early life maternal deprivation (MD) and chronic unpredictable stress (CUS) in adulthood on NTGinduced migraine-like symptoms in rats. In response to NTG, increased head-scratching and facial rubbing, and decreased climbing behavior were associated with more severe migraine attacks in female rats. Moreover, NTG-related head-scratching and facial rubbing scores were aggravated in female but not male rats subjected to either CUS or MD stress. Body-grooming behavior was significantly increased in stressed male rats. Chronic unpredictable stress experienced in adulthood subsequent to MD did not change NTG-induced migraine symptoms in rats of either sex.
NTG effectively models migraine-like symptoms in rodents. A single or repeated administration of NTG to the dura mater, or its systemic application, mimics various migraine-associated key features, including hyperalgesia, allodynia, and other nociceptive-specific behaviors, such as head and face scratching and reduced physical activity [5,39]. Potentially, NTG produces a neurovascular component in spinal and second-order trigeminal neurons, resulting in peripheral and central sensitization by induction of pro-nociceptive and pro-inflammatory chemical mediators in brain stem areas involved in migraine pathophysiology [11]. In patients who suffer from migraine without aura, orally administered NTG exhibited a significant heat pain threshold reduction. The patients experienced a slight headache immediately after NTG intake and a moderate or severe subsequent headache accompanied by non-headache symptoms like nausea, vomiting, and photophobia [11].
In line with previous studies, our data shows that female rats manifest more severe NTG-induced migraine symptoms. Intradermal administration of NTG displayed greater thermal hypersensitivity in female rats, which was relieved by sumatriptan as an anti-migraine drug [4]. Moreover, intravenous application of NTG in female rats provoked the expression of c-fos protein in several brain nuclei involved in migraine pain modulation [15].
The mechanism(s) underlying sex differences in migraine headaches are complex and not fully understood. It has been indicated that menstrual-related changes in sex hormones play a pivotal role in the onset, duration, and severity of migraine attacks [36]. Drops in estrogen levels can induce migraine, and diminishing this failure may inhibit the attacks [8]. Furthermore, neuroimaging studies have reported obvious functional and anatomical differences between migraine sufferers of both genders. For example, female migraineurs have thicker posterior insula and precuneus cortices than male ones [28]. NTG-induced migraine has been shown to decrease the expression of signaling molecules including BDNF, phosphorylated extracellular signal-regulated kinase (p-ERK), and phosphorylated c-AMP-responsive element binding protein. (p-CREB) in the brains of female ovariectomized rats. Such effects have been reserved with exogenous estrogen treatment [42]. Although this study did not evaluate the mechanism(s) that explain why females are more sensitive to migraine than males, it could be related to fundamental neuroanatomical and hormonal gender differences.
The present study results also show that NTG-induced migraine-like symptoms were exaggerated in maternally deprived female rats, while male rats were not affected. To the best of our knowledge, this study is the first to investigate the MD stress and migraine interaction considering gender differences. However, it has been previously reported that MD induces complicated gender-dependent alterations in pain processing that persist into adulthood [25]. Subsequent to a peripheral nerve injury, female rats exposed to MD demonstrate more hyperalgesia, mechanical allodynia, and neuro-inflammation compared to male rats [6]. Maternal stress could increase formalin-induced pain, which was more pronounced in female rats compared to males [7]. In contrast, Prusator et al., reported that male rats showed enhanced visceral and somatic pain sensitivity due to MD stress, whereas female rats did not [35].
There is a narrow cross-regulatory association between the hypothalamic-pituitaryadrenal axis (HPA) and the hypothalamic-pituitary-gonadal (HPG) axis that allows one axis to modulate another. Various studies have shown that perturbations in the HPA axis activity exacerbate the HPG axis coordination [2,16]. A hyperactive HPA axis during chronic stress has been considered as a primary mechanism underlying reproductive dysfunction in human and animal studies [37]. Gonadal hormone production deficiency has also been shown to have an influence on the HPA axis dynamics [2,32]. The prolonged stressors change the circulating levels of glucocorticoids and their balance with HPG axis hormones, including testosterone and estradiol [29,41]. Certainly, sexrelated stress-induced changes in the HPA and HPG axis, some chemical and physiological differences in stress control brain regions, and their associated endocrine rhythms might be accompanied by different behavioral and physiological responses in males and females. However, there is still limited scientific knowledge about how the axis coordinates their hormonal harmony to modulate migraine pain perception.
The maturation of the pain modulatory systems may be affected by pituitary-adrenal hormones [30]. There have been extensive studies conducted on the relationship between corticosteroids and altered pain sensitivity, however, few studies have examined the interplay of gender and nociception. As a limitation, in this present study, sex hormones were not considered. This area may be a good opportunity to understand the underlying mechanism of gender-specific features of the relationship between MD and migraine headache.
An unexpected result of the present study was that NTG-induced pain symptoms were not exaggerated in rats exposed to a combination of MD and CUS compared to both male and female MD-only rats. One interpretation of this finding could be that migraine pain is not affected by CUS, which might to some extent be associated with the heterogeneous effects of CUS on pain modulation. Based on the experimental conditions, such as those used in this experiment, the duration and severity of nociception, and the age of the animals, CUS may exacerbate or inhibit nociceptive responses [20,34,35]. Using an NTG model of migraine, Kaufmann and Brennan in their 2018 study, indicated that either chronic variable stress or social defeat stress was not able to alter mechanical withdrawal thresholds in male migrainous mice. NTG effects were more quickly reduced in mice exposed to chronic variable stress than in control and social defeat stressed groups [20]. The CUS paradigm would have substantial physiological components that limit its translational relevance. Most clinical scenarios do not involve such robust physiological stressors. In this study, it seems likely that the nature of the stressors is likely to be more sensitive to migraine vulnerability than increased stress levels. Additional work using other models of chronic stress is essential to endorse the CUS effects on migraine characterization.
The accumulation of life stress is associated with enhanced neurological disorders in adulthood. However, negative childhood experiences may promote adaptive capacity features to build resilience to stressful challenges in later life [10]. Stress resilience in rodents has been shown to lead to increases in pain threshold and decreases in stressinduced hyperalgesia [3,13]. Interestingly, NTG-induced pain sensitivity has been decreased in stress-resilient mice compared to stresssensitive mice [21]. It is therefore possible that exposure to one stress paradigm in early life may lead to adaptive stress responses to cope with advanced stressors. However, it is still unclear how such stress programming can ensure specific migraine resilience profiles in adulthood.
The assessment of rodent grooming behavior is useful for the evaluation of repetitive behavior in studies involving pain and/or stress [19,40]. In this study, body-grooming behavior was significantly decreased in male rats with migraine, but not in female rats subjected to either CUS or MD stress. Potentially, there could be some differences between the sexes in sensitivity to stress and clinical exhibition of migraine states. More research work is needed to define grooming behavior related to migraine.
Consistent with previous studies and to avoid unnecessary animal suffering, we used 7 rats in each study group. This relatively small sample size might lead to negative behavioral scales given the behavioral outcomes targeted.

Conclusions
These results suggest that maternal deprivation stress enhancesvulnerability to NTG-induced migraine headaches during adulthood in female but not male rats. However, maternally deprived male or female rats subjected to later chronic unpredictable stress did not show any additional migraine-aggravating behavior.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.