Neonatal Maternal Deprivation Followed by Adult Stress Enhances Adrenergic Signaling to Advance Visceral Hypersensitivity

The pathophysiology of visceral pain in patients with irritable bowel syndrome remains largely unknown. Our previous study showed that neonatal maternal deprivation (NMD) does not induce visceral hypersensitivity at the age of 6 weeks in rats. The aim of this study was to determine whether NMD followed by adult stress at the age of 6 weeks induces visceral pain in rats and to investigate the roles of adrenergic signaling in visceral pain. Here we showed that NMD rats exhibited visceral hypersensitivity 6 h and 24 h after the termination of adult multiple stressors (AMSs). The plasma level of norepinephrine was significantly increased in NMD rats after AMSs. Whole-cell patch-clamp recording showed that the excitability of dorsal root ganglion (DRG) neurons from NMD rats with AMSs was remarkably increased. The expression of β2 adrenergic receptors at the protein and mRNA levels was markedly higher in NMD rats with AMSs than in rats with NMD alone. Inhibition of β2 adrenergic receptors with propranolol or butoxamine enhanced the colorectal distention threshold and application of butoxamine also reversed the enhanced hypersensitivity of DRG neurons. Overall, our data demonstrate that AMS induces visceral hypersensitivity in NMD rats, in part due to enhanced NE-β2 adrenergic signaling in DRGs.


Introduction
Irritable bowel syndrome (IBS) is a common gastrointestinal disease characterized by disorders of intestinal motility and accompanied by chronic abdominal pain [1][2][3]. The treatment of chronic abdominal pain is difficult [4]. Research on this disease is progressing slowly due to the lack of suitable animal models. Although the currently available research provides insights into the processing and regulation of chronic pain [5,6], the precise pathophysiology of IBS has not been fully elucidated and effective strategies for treating the primary symptoms are not available [1,7,8]. Our previous studies have shown that neonatal maternal deprivation (NMD) can induce chronic visceral pain in adult rats at the age of 7 weeks but not at 6 weeks [9,10]. It seems that the age of 7 weeks is an important time point for the development of visceral hypersensitivity in rats with NMD. However, it is unknown whether 6-week-old NMD rats are more sensitive to environmental stimuli than age-matched controls.
A growing body of evidence has demonstrated that severe adverse environmental factors such as stress might be a stimulus to generate visceral hypersensitivity at the adult age in humans [11,12] and animals [13][14][15]. For example, repetitive water-avoidance stressors for 10 days induce visceral hypersensitivity in rats and mice [16]. Heterotypic intermittent stress for 9 consecutive days induces visceral hypersensitivity immediately after Wan-Jie Du and Shufen Hu have contributed equally to this work.
& Guang-Yin Xu guangyinxu@suda.edu.cn termination of the last stressors [14,17]. However, single or mild stressors do not induce visceral hypersensitivity in healthy adult animals. In the present study, we designed a stress protocol to determine whether one stressor or mild stressors could induce visceral pain in NMD rats at the age of 6 weeks. Adrenergic signaling plays many important roles in the nervous system to regulate stress responses [18][19][20]. Adrenergic receptors (ARs) are classically divided into two main groups: aand b-adrenoceptors. The a-adrenoceptors include a 1A , a 1B , a 1D , a 2A , a 2B , and a 2C subtypes, and b-adrenoceptors into b 1 , b 2 , and b 3 subtypes [21]. They have been reported to be expressed in primary sensory neurons with their cell bodies located in the dorsal root ganglia (DRGs) [22,23]. Previous studies have suggested that the b 1 and b 2 subtypes are involved in the adrenergic activation [24] that may play a role in colonic transit. The b 2 ARs are reported to produce a hyperalgesic state in rats [22,25,26]. The b 3 ARs, mainly expressed in brown and white adipose tissue, regulate energy metabolism and thermogenesis [27]. Besides, other studies have shown that the adrenergic system plays a role in the visceral pain caused by chronic stress [13]. Recently, we have reported that adrenergic b 2 receptors mediate visceral pain evoked by heterotypic intermittent stress in rats [14]. However, whether adrenergic signaling in primary sensory neurons participates in visceral pain of NMD rats at the age of 6 weeks after additional adult stress is unknown.
Thus, we designed this study to test the hypothesis that adrenergic activation plays a crucial role in the switch of NMD rats at the age of 6 weeks from no pain status to pain hypersensitivity induced by stress in adulthood.

Animals
We used male Sprague-Dawley rats in the present experiments. The experimental protocol was approved by the Institutional Animal Care and Use Committee of Soochow University. Animal care and handling were strictly in accordance with the regulations and guidelines of the International Association for the Study of Pain.

Adult Stress Protocol to Induce Visceral Pain
The stress protocol to induce chronic visceral hyperalgesia is shown in Fig. 1. This protocol had two parts, NMD and stress in adulthood. The adult stress was divided into an adult single stressor (ASS), cold-restraint stress (CRS), for 45 min, and adult multiple stressors (AMSs) consisting of CRS for 45 min, forced-swimming stress (FSS) for 20 min, and water-avoidance stressor (WAS) for 60 min. The interval between each stressor was 60 min. NMD was imposed as described previously [9,10,28]. In brief, pups were separated from the maternity cages and placed in different cages with an electric blanket to maintain body temperature at *32°C for 3 h every day from postnatal days 2 to 15. After the 3 h of separation, pups were returned to the dam cages. Littermates in the control group were not handled and were kept in the maternity cages with their dam. Both groups were exposed to an ASS or AMSs on postnatal day 42 (Fig. 1). All experiments were performed 6 h or 24 h after termination of last stressor unless indicated otherwise. Multiple batches of rats at the age of 6 weeks were used in the present study. The numbers of animals in each group were as follows: control group, 15; NMD group, 17; CON ? ASS group, 13; NMD ? ASS group, 16; CON ? AMS group, 38; and NMD ? AMS group, 68.

Measurement of Behavioral Responses to Graded Colorectal Distention (CRD)
Chronic visceral hyperalgesia was assessed by grading the behavioral response of rats to CRD at the age of 6 weeks based on previous publications [9,10,20]. All behavioral tests were performed in a blinded manner.

Drug Administration
In the behavioral experiments, 5 mg/kg butoxamine (BUTO, a b 2 antagonist; Sigma, St. Louis, MO), 3 mg/kg propranolol (PROP, a norepinephrine b receptor antagonist; Sigma) or 3 mg/kg phentolamine (PHEN, an a receptor antagonist; Sigma) dissolved in 0.9% normal saline (NS) was intraperitoneally injected into AMS rats once for behavioral experiments and once daily for 7 consecutive days for patch-clamp recordings and western blotting. The drug concentrations used were based on our previous study and reports from other groups [14,20].

Measurement of Norepinephrine (NE) in Blood Plasma
Blood samples were collected from the trunk into centrifuge tubes containing 0.45% citric acid and 2.5% sodium citrate at euthanasia by decapitation. After refrigerated centrifugation, the supernatant was quickly aliquoted and stored at -80°C for experiments. NE levels in the plasma were determined using an enzyme immunoassay kit from Abnova (Norepinephrine ELISA Kit), as previously described [20].

Dissociation of DRG Neurons and Whole-Cell Patch-Clamp Recordings
Rats from AMS-treated NMD or control (*6 weeks) were sacrificed by decapitation. The detailed procedures for the acute isolation of DRG neurons and patch clamp recordings were as previously reported [14,20,31]  HEPES, 10 glucose, 5 EGTA, 1 CaCl 2 , pH 7.25 adjusted with KOH; osmolarity 292 mOsm.

Data Analysis
All data are presented as mean ± SEM. Statistical testing was performed using OriginPro 8 (OriginLab, Northampton, MA). Normality was first checked for all data before analysis. Significance was determined using the twosample t test, Mann-Whitney test, Mann-Whitney test following Friedman ANOVA, or Tukey's post-hoc test following two-way repeated measures ANOVA. P \ 0.05 was considered to be statistically significant.

AMS Increases NE Concentration in Blood Plasma of NMD Rats
NE is one of the important molecules involved in the regulation of stress responses [18,20,32]. In the present experiments, we demonstrated that there was no significant difference in NE concentrations in the plasma between control and NMD groups (P [ 0.05, two-sample t test, Fig. 3A). There was also no significant difference in NE concentration in the plasma between the CON ? ASS and NMD ? ASS groups (P [ 0.05, two-sample t test, Fig. 3B). However, there was a significant difference in NE concentrations in the plasma between CON ? AMS and NMD ? AMS rats, indicating that AMS remarkably increased the NE levels (P \ 0.01, two-sample t test, Fig. 3C).

AMS Enhances Excitability of Colon DRG Neurons in NMD Rats
To determine the effects of NMD ? AMS on the excitability of colon DRG neurons, whole-cell patch clamp recordings were carried out. The resting membrane potentials of T 13 -L 2 DRG neurons were -49.25 ± 1.92 mV (n = 8 cells) and -46.3 ± 1.59 mV (n = 10 cells) for CON ? AMS and NMD ? AMS rats, respectively (Fig. 4A). After statistical analysis, there was no significant difference in resting membrane potentials between CON ? AMS and NMD ? AMS rats (P [ 0.05, two-sample t test). Rheobase and firing patterns in response to current stimulation were also recorded. The average rheobases of colon DRG neurons were 56 ± 9.6 pA (n = 8 cells) from CON ? AMS rats and 13 ± 2.1 pA (n = 10 cells) from the corresponding NMD ? AMS rats (Fig. 4B, P \ 0.01, Mann-Whitney test). The number of action potentials (APs) in response to 29 rheobase current stimulation was significantly increased in DRG neurons from the NMD ? AMS group ( Fig. 4C and D, P \ 0.05, twosample t test). However there was no significant difference in the number of APs evoked by 39 rheobase current stimulation ( Fig. 4C and D, P[0.05, two-sample t test). In addition, we counted the numbers of APs induced by 100 pA, 300 pA and 500 pA ramp current stimulation (Fig. 4E and F). The numbers of APs in response to 100 pA, 300 pA, and 500 pA ramp current stimulation differed significantly between CON ? AMS and NMD ? AMS rats (P \ 0.05, P \ 0.001, two-sample t test). At the same time, the latency of APs evoked by 100 pA, 300 pA, and 500 pA ramp current stimulation was significantly lower in the NMD ? AMS group than in the CON ? AMS group (Fig. 4G, P\0.01, Mann-Whitney test), indicating that the neuronal excitability was enhanced in NMD rats after AMS treatment.

Treatment with b 2 Receptor Antagonist Reverses Visceral Hypersensitivity
Next we determined whether AR antagonists can reverse chronic visceral hyperalgesia. Injection of PROP, a nonselective b receptor antagonist, significantly reversed visceral hyperalgesia in a time-dependent manner (Fig. 6A, NS: n = 12; PROP: n = 3, P \ 0.001, compared with NS, Tukey post-hoc test following two-way repeated measures ANOVA). Further, injection of BUTO, a b 2 receptor antagonist, significantly raised the DT of NMD ? AMS rats (Fig. 6B, n = 12 rats/group, P \ 0.05, P \ 0.001, Tukey post-hoc test following two-way repeated measures ANOVA). However, injection of PHEN, an antagonist of a receptors, had no effect on the visceral hyperalgesia induced by AMS (Fig. 6C, NS: n = 12; PHEN: n = 3, P [ 0.05, Tukey post-hoc test following two-way repeated measures ANOVA). In addition, there was no significant effect of BUTO at the same dose on the DT of CON ? AMS rats (Fig. 6D, n = 5 rat/group, P [ 0.05, Tukey posthoc test following two-way repeated measures ANOVA). These data suggested that b 2 receptors are involved in the development of the visceral hypersensitivity induced by AMS in NMD rats.

Treatment with b 2 Receptor Antagonist Reverses Neuronal Hyperexcitability
To investigate the role of b 2 receptors in NMD ? AMS rats, we injected the b 2 receptor antagonist BUTO once daily for 7 consecutive days to assess its influence on the excitability of colonic T 13 -L 2 DRG neurons. The resting membrane potentials were -40.78 ± 0.38 mV (n = 9 cells) in NS rats and -49.23 ± 1.11 mV (n = 13 cells) in BUTO rats (Fig. 7A, P \ 0.001, Mann-Whitney test). The average rheobases were 18 ± 2 pA (n = 9 cells) in NS rats and 33 ± 4 pA (n = 13 cells) in BUTO rats (Fig. 7B, P \ 0.01, Mann-Whitney test). The numbers of APs in response to 29 and 39 rheobase current stimulation were significantly lower in DRG neurons after BUTO treatment (Fig. 7C, D, P \ 0.05, P \ 0.001, Mann-Whitney test and two-sample t test). In addition, we counted the number and measured the latency of APs induced by 100 pA, 300 pA, and 500 pA ramp current stimulation (Fig. 7E-G). The numbers of APs were significantly decreased and their D NMD ? AMS significantly enhanced the mRNA level of b 2 receptors when compared with CON ? AMS groups. E NMD did not alter the expression of b 2 receptors when compared with CON groups at 6 weeks. F NMD ? ASS significantly enhanced the expression of b 2 receptors when compared with CON ? ASS groups. *P \ 0.05, **P \ 0.01, two-sample t test.
latency was markedly increased (P \ 0.001, Mann-Whitney test and two-sample t test). These results demonstrated that BUTO decreased the neuronal excitability enhanced by NMD and AMS exposure.

Discussion
In the present experiments, we demonstrated that NMD rats exhibited reduced thresholds and increased abdominal withdrawal reflex scores to CRD when compared with age-matched control rats after exposure to multiple stressors as adults. This indicates that a combination of NMD and AMS exacerbated the symptoms by enhancing visceral hypersensitivity in NMD rats at the age of 6 weeks. This also supports an idea that NMD puts such rats at risk when they grew to the age of 6 weeks although these rats do not show any visceral hypersensitivity as reported previously [31]. Although these rats are more sensitive to multiple stressors at the age of 6 weeks than age-matched control rats, this does not mean that these NMD rats are sensitive to any environmental stimulus, since the CRS alone did not induce any visceral response. Environmental stimuli have to reach a minimal threshold to induce visceral hypersensitivity. Although we did not define the minimal threshold for NMD rats, our data might have clinic relevance in that an adverse neonatal stimulus followed by adult stressors aggravates the symptoms of IBS or/and shortens the time window to induce visceral hypersensitivity. The finding of enhanced visceral sensitivity was strongly supported by the enhanced neuronal excitability in NMD rats followed by AMS. By whole-cell patch clamp recording, we showed that the excitability of DRG neurons was remarkably enhanced in NMD rats followed by AMS when compared with that of age-matched control rats followed by AMS. Our electrophysiological data provide a cellular mechanism underlying the enhanced visceral pain behaviors. Of note is that the ionic basis for this enhanced cellular excitability needs further investigation.
In the present experiments, we focused on the mechanism by which AMS produced visceral hypersensitivity and neuronal hyperexcitability in NMD rats at the age of 6 time-dependent manner. C The a receptor antagonist phentolamine (PHEN, 3 mg/kg) had no effect on the visceral hyperalgesia induced by NMD ? AMS. D BUTO (5 mg/kg) had no effect on CON ? AMS rats. **P \ 0.01, ***P \ 0.001, Tukey post-hoc test following twoway repeated measures ANOVA. Note that the NS group in panels A-C used the same rats.
weeks. A recent study has shown that chronic stress involves NE release, AR expression, and/or the activation of intermediates in AR-induced signaling, thus contributing to the pathology of many immune-mediated diseases [33]. Adrenergic activation is involved in neuropathic and inflammatory pain states [34,35]. Our previous studies showed that neonatal colonic inflammation or heterotypic intermittent stress increases the NE concentration in blood plasma without alterations in b 2 AR expression in DRGs [14,20]. A new finding in the present study was that AMS not only enhanced the plasma NE levels but also increased the expression of b 2 ARs in DRGs at 6 weeks. This discrepancy might be due to the different stimulus protocols used. The neonatal colonic inflammation model was established only by one colonic stimulus in the neonate.
The heterotypic intermittent stress model was induced by stressors only in adulthood [14,17]. However, the present model was established by a combined stress protocol in both neonates and adults. This combination of stimuli might be more relevant to the clinic situation. Therefore, our study also provides a good animal model to better mimic the clinic situation in patients with IBS, thus providing a better basis on which to investigate the mechanisms of visceral pain. We showed that b 2 ARs played an important role in the peripheral nervous system. The expression of b 2 ARs at the protein and mRNA levels was remarkably upregulated in NMD rats with AMS while the expression of b 1 and b 3 ARs was not altered greatly. Inhibition of b 2 ARs by PROP or BUTO enhanced the CRD threshold in a time-dependent manner while inhibition of a adrenergic receptors by PHEN did not affect the CRD threshold. Furthermore, application of BUTO also reversed the enhanced hypersensitivity of DRG neurons. This anti-nociceptive effect was specific since BUTO did not have any effect on agematched control rats with AMS. These data demonstrated that enhanced NE and b 2 adrenergic signaling plays an important role in increasing the visceral hypersensitivity in rats with NMD and AMS. Of note is that the protein expression of b 2 ARs in NMD rats followed by ASS was also significantly increased. However, the NE level in the plasma was not altered remarkably. This may explain why the NMD rats followed by ASS did not show enhanced visceral sensitivity. In addition, we do not know whether *** AMS can exacerbate the visceral hypersensitivity of NMD rats at the age of 7 weeks, although NMD rats at this age already exhibit enhanced visceral hypersensitivity. In the present study, we only focused on the peripheral mechanisms; the central mechanisms, such as spinal synaptic plasticity [36], deserve further investigation.
In summary, we demonstrated that NMD followed by AMS increased visceral hypersensitivity in association with an elevation of the NE concentration in plasma, the expression of b 2 ARs in DRGs, and the neuronal excitability of colonic DRG neurons. Blockade of b 2 ARs attenuated visceral hypersensitivity to colorectal distension and neuronal hyperexcitability. Together with our previous studies, our data provide additional evidence to support the idea that the NE-b 2 signaling pathway plays an important role in the development of visceral hypersensitivity. This study might shed light on the pathogenesis of visceral hypersensitivity imposed by environmental stress during early and adult life. AR inhibitors might serve as alternates to relieve abdominal pain in patients with IBS.