The authors have declared that no competing interests exist.
Conceived and designed the experiments: ZW DPH EAM SB. Performed the experiments: ZW MAO RDP DPH. Analyzed the data: ZW MB DPH. Contributed reagents/materials/analysis tools: ZW MB DPH. Wrote the paper: ZW DPH EAM MB.
Repeated water avoidance stress (WAS) induces sustained visceral hyperalgesia (VH) in rats measured as enhanced visceromotor response to colorectal distension (CRD). This model incorporates two characteristic features of human irritable bowel syndrome (IBS), VH and a prominent role of stress in the onset and exacerbation of IBS symptoms. Little is known regarding central mechanisms underlying the stress-induced VH. Here, we applied an autoradiographic perfusion method to map regional and network-level neural correlates of VH. Adult male rats were exposed to WAS or sham treatment for 1 hour/day for 10 days. The visceromotor response was measured before and after the treatment. Cerebral blood flow (CBF) mapping was performed by intravenous injection of radiotracer ([14C]-iodoantipyrine) while the rat was receiving a 60-mmHg CRD or no distension. Regional CBF-related tissue radioactivity was quantified in autoradiographic images of brain slices and analyzed in 3-dimensionally reconstructed brains with statistical parametric mapping. Compared to sham rats, stressed rats showed VH in association with greater CRD-evoked activation in the insular cortex, amygdala, and hypothalamus, but reduced activation in the prelimbic area (PrL) of prefrontal cortex. We constrained results of seed correlation analysis by known structural connectivity of the PrL to generate structurally linked functional connectivity (SLFC) of the PrL. Dramatic differences in the SLFC of PrL were noted between stressed and sham rats under distension. In particular, sham rats showed negative correlation between the PrL and amygdala, which was absent in stressed rats. The altered pattern of functional brain activation is in general agreement with that observed in IBS patients in human brain imaging studies, providing further support for the face and construct validity of the WAS model for IBS. The absence of prefrontal cortex-amygdala anticorrelation in stressed rats is consistent with the notion that impaired corticolimbic modulation acts as a central mechanism underlying stress-induced VH.
Considerable evidence links stress with the onset and symptom exacerbation in irritable bowel syndrome (IBS)
There are few published reports on functional brain mapping studies in stress-induced visceral hyperalgesia animal models. Stam et al.
We have recently adapted an autoradiographic cerebral blood flow (CBF) perfusion mapping method to the rat CRD model
Repeated, daily WAS (7–10 days) induces a chronic visceral hyperalgesia in the rodent model. This hyperalgesia persists for periods as long as one month after cessation of the stress
One major limitation of FC analysis is that correlation based analysis does not address causality. Furthermore, due to multiple comparisons, false positive findings are inevitable when a simple significance threshold is applied. Constraining FC analysis with structural connectivity (SC) information can reduce the number of false positive reports, as well as provide directionality for the otherwise non-directional FC networks. The concept of anatomically constrained FC analysis has been implemented in effective connectivity analysis of human and animal brain imaging data
Adult male Wistar rats (2–2.5 month old) were purchased from Harlan Sprague Dawley (Indianapolis, IN, USA) and were individually housed in the vivarium on a 12-hour light/12-hour dark cycle with free access to water and rodent chow. All experiments were conducted under a protocol approved by the Institutional Animal Care and Use Committee of the University of Southern California, an institution accredited by the Association for Assessment and Accreditation of Laboratory Animal Care, International. All work was in accordance with the guidelines of the Committee for Research and Ethical Issues of the International Association or the Study of Pain. The numbers of animals in each group were as follows: Sham stress/0-mmHg CRD, n = 9; Sham stress/60-mmHg CRD, n = 10; WAS/0-mmHg CRD, n = 10; WAS/60-mmHg CRD, n = 10.
One week before the start of WAS treatment (
Visceral motor response (VMR) to colorectal distension (CRD) was measured before (day 0, baseline) and after (day 11) water avoidance stress (WAS) or sham treatment. Each time, CRDs of 10-, 20-, 40-, and 60-mmHg (duration = 20 s, interstimulus interval = 4 min) were delivered twice for each pressure level. On day 11, following VMR measurement, cerebral blood flow (CBF) mapping was performed while the animal was receiving a 60-mmHg CRD or no distension (0-mmHg control).
VMR to CRD was assessed as described before
The protocol was as described before
On day 11, animals were allowed to rest for 15 min following the last distension of the CRD series. A piece of silastic tubing was filled with radiotracer [14C]-iodoantipyrine (125 µCi/kg in 300 µL of 0.9% saline, American Radiolabelled Chemicals, St. Louis, MO, USA). The radiotracer-filled tubing was then connected to the animal's cannula on one end, and to a syringe filled with euthanasia agent (pentobabital 50 mg/mL, 3 M potassium chloride) on the other. The animal was allowed to rest for another 5 min before receiving one episode of 60-mmHg. Thirty-five seconds after the onset of the distension, radiotracer was infused at 2.25 mL/min by a motorized pump, followed immediately by 0.7 mL of euthanasia solution, which resulted in cardiac arrest within ∼10 s, a precipitous fall of arterial blood pressure, termination of brain perfusion, and death
Brains were rapidly removed, flash frozen in methylbutane on dry ice (∼−55°C) and embedded in optimal cutting temperature compound (Sakura Finetek Inc., Torrance, CA,USA). Brains were subsequently sectioned on a cryostat (HM550 series, Microm International GmbH, Walldorf, Germany) at −18°C into 20-µm thick coronal slices, with an inter-slice spacing of 300 µm. Slices were heat-dried on glass slides and exposed to Kodak Biomax MR films (Eastman Kodak, Rochester, NY, USA) for 3 days at room temperature. Images of brain sections were then digitized on an 8-bit gray scale using a voltage stabilized light box (Northern Lights Illuminator, Interfocus Imaging Ltd., Cambridge, UK) and a Retiga 4000R charge-coupled device monochrome camera (Qimaging, Surrey, Canada). Autoradiographic CBF mapping in rodents has a spatial resolution of 100-µm, and hence, can provide information on sub-regional activation, such as in individual amygdaloid nuclei.
rCBF-related tissue radioactivity was quantified by autoradiography and analyzed on a whole-brain basis using statistical parametric mapping (SPM, version 5, Wellcome Centre for Neuroimaging, University College London, London, UK). Recently, we and others have developed and validated an adaptation of SPM for use in rodent brain autoradiograph
To test the hypothesis that WAS may result in altered corticolimbic modulation during noxious visceral stimulation, we applied seed-region correlation analysis to assess differences in the CRD-evoked FC of the PrL of the PFC between treatment groups. The seed region of interest (ROI) was hand drawn in MRIcro (version 1.40,
Anatomical (structural) connectivity of the PrL in the rat was extracted from BAMS. BAMS includes a large set of rat structural connections collated from the literature, or directly inserted by neuroanatomists
Results of the SPM seed correlation analysis were only analyzed for those regions structurally connected with the PrL. This is equivalent to taking an intersection of the structural and functional connectivity network of the PrL, resulting in an SLFC network (
Structural (SC) and functional connectivity (FC) networks are combined to create a structurally linked functional connectivity (SLFC) network such that final network contains all regions that the SC and FC networks have in common. Note the SLFC network inherits directionality information (denoted with arrows) from the SC network and the sign of functional interaction (positive/red or negative/blue correlation) from the FC network.
Ten days of WAS induced significant increases in visceromotor response to CRD on day 11 as compared to day 0 baseline (
(A) VMR measured as electromyographic (EMG) area under the curve (AUC) and expressed as % of control (baseline VMR to 60-mmHg CRD) was significantly increased following WAS on day 11 compared to day 0 baseline (
CRD-evoked functional brain activation was assessed for each treatment type by contrasting the subgroup receiving 60-mmHg CRD, and the one receiving no CRD (0-mmHg control) of the same treatment type (
Statistically significant increases (red scale) and decreases (blue scale) in regional cerebral blood flow (rCBF) contrasting the subgroup receiving 60-mmHg CRD and the one receiving no distension (0-mmHg) are shown for the sham (left column) and water avoidance stress (WAS)-treated rats (middle column) over representative coronal slices of the brain template with anterior-posterior coordinates given relative to the bregma (
In contrast, WAS-treated rats showed CRD-evoked brain activation in a drastically different pattern. Major differences were noted in the magnitude and extent of regional activation. Greater activation in the WAS rats was noted in the anterior and posterior insula, and in the amygdala (central n., lateral n., basolateral n., basomedial n., medial n., bed nucleus of the stria terminalis intraamygdaloid division). The WAS rats also showed significant activation in the hypothalamus (medial preoptic area, medial preoptic n., lateral preoptic area, ventromedial hypothalamic n.), nucleus accumbens, and bed nucleus of the stria terminalis (medial division), which was not seen in the sham rats. Whereas sham rats showed activation in the anterior dorsal aspect of caudate putamen, WAS rats showed activation in the posterior and anterior ventral aspects of caudate putamen. Importantly,
Seed correlation-based FC analysis was constrained by SC information of the PrL to generate SLFC of the PrL cortex (
The SLFC network contains both directionality information of the underlying structural connections (denoted with arrows) and the sign of functional interaction (positive/red or negative/blue statistically significant correlation).
1. Positive FC with a cluster of cortical regions over largely bidirectional structural connections, including orbital (ventral), secondary motor, cingulate, retrosplenial (dysgranular, granular), anterior insular, and ectorhinal cortices (PrL[+]←←Ctx).
2. Positive FC over projections to the striatum (PrL[+]←Striatum) or from lateral orbital cortex (PrL[+]←LO).
3. Negative FC over afferent projections from the hippocampal formation (CA1, dorsolateral entorhinal cortex) (PrL[-]←Hippocampal formation).
4. Negative FC with the hypothalamus over efferent projections to medial preoptic nucleus, the anterior hypothalamic area (anterior part), and subventricular zone (PrL[-]←Hypothalamus), and afferent projections from anterior hypothalamic (central part) and supramammillary nucleus (PrL[-]←Hypothalamus).
5. Positive FC with the thalamus over efferent projections to the ventral anterior, ventral lateral, anterior ventral, and reticular nuclei, and medial and lateral habenular nuclei (PrL[+]←Thalamus), as well as bidirectional connections with medial dorsal and central nuclei (PrL[+]←←Thalamus).
6. Negative FC over efferent projection to the dorsomedial and dorsolateral periaqueductal gray in the midbrain (PrL[-]←PAG).
WAS-induced changes in SLFC were most noticeable in relation to a cluster of cortical regions and to the amygdala. Whereas PrL in the sham animals showed significant, positive FC with a cortical cluster consisting of secondary motor, dorsal posterior cingulate (pCg1), ventral cingulate (Cg2), retrosplenial (dysgranular, granular) cortices, this connectivity turned
In addition, WAS-induced alterations in SLFC of the PrL/PFC included the following:
1. Changes from positive to negative FC with the thalamus (ventral anterior, ventral lateral, anterior ventral, mediodorsal, centromedial, medial and lateral habenular nuclei) (PrL[-]→/←/← →Thalamus), subthalamic nucleus, and zona incerta.
2. Changes from negative to positive FC with the hypothalamus (medial preoptic n.; anterior hypothalamic area, anterior and central parts; subventricular zone) (PrL[+]→/←Hypothalamus).
3. New (no FC in sham rats) positive FC with infralimbic, posterior insular, and piriform cortices (PrL[+]→/← →Ctx), hypothalamus (medial mammillary n., ventromedial n., lateral preoptic area)(PrL[+]←/← →Hypothalamus), dorsal subiculum, nucleus accumbens, claustrum, dorsal endopiriform nucleus, ventral pallidum, and nucleus of the diagonal band.
4. New, negative FC with thalamus (posterior, reuinens, rhomboid, centrolateral nuclei)(PrL[-]→/←/← →Thalamus), and brainstem areas (substantia nigra reticulata and compacta, isthmic and mesencephalic reticular formation, lateral and ventrolateral periaqueductal gray, dorsal raphe n., dorsal and laterodorsal tegmental n., pontine n., central gray, subcoeruleus n. alpha)(PrL[-]→/← →Brainstem).
Neurobiological sequelae of chronic stress have been the subject of extensive research. For example, it has been well established that chronic stress in rodents can induce both visceral
In the current study, stressed compared to sham treated rats showed greater CRD-evoked activation broadly across the amygdala, with the central amygdaloid nucleus showing the most significant differences. These findings are consistent with an extensive literature on stress induced sensitization of the amygdala. For example, the amygdala has been implicated in chronic stress-induced sensitization of anxiety- and fear-related responses to an acute stressor
Activation of the PrL during acute visceral pain in nonstressed rats has previously been reported
Functional interaction between brain regions can be analyzed through FC analysis of brain imaging data. Here, we focused FC analysis on the PrL and constrained FC results with SC information of the PrL. The integration of whole brain-level FC and SC information reflects a recent trend in human brain imaging field to take advantage of rapid advances in the Human Connectome Project
In rodents, the PrL is reciprocally connected with the lateral and basolateral nuclei of the amgydala, and sends efferent projections to the central nucleus and anterior part of basomedial nucleus, and receives afferent projections from medial, posterior, and cortical nuclei and posterior part of the basomedial nucleus of the amygdala. In sham rats, PrL/PFC SLFC during CRD was characterized in our study by negative FC with the amygdala nuclei over mostly afferent projections from the amygdala (PrL[-]←Amygdala), except bidirectional connection with the lateral nucleus (PrL[-]← →La). These SLFC results suggest that in sham animals, PrL/PFC inhibits the amygdala through its projection to the lateral amygdalar nucleus, whereas the amygdala may in return inhibit PrL/PFC activity though its lateral, basomedial, medial, and cortical nuclei.
Bidirectional PFC-amygdala interactions have been extensively studied in humans and in laboratory animals, and changes in these interactions have been implicated in the regulation of negative emotion, mood and pain
In the current study, stressed rats compared to sham rats showed substantial changes in the SLFC of PrL/PFC. The negative FC with the amygdala seen in the sham disappeared, and in its place appeared a few positive connections (with the lateral and corticoamygdaloid nuclei). Our results are consistent with those of Correll et al.
Previous imaging studies in animal and human subjects [reviewed by 61] have also implicated the anterior cingulate cortex in visceral pain processing and regulation, with the majority of visceral distension studies reporting enhanced activation of mid-cingulate subregions. Here, the observed deactivation in the cingulate area in the stressed rats and no activation in the sham rats was unexpected. Previously, we have reported cingulate activation in male, naïve rats (with no prior experience of CRD) receiving acute, noxious CRD
Whereas both sham and stressed rats showed CRD-evoked activation in the anterior and posterior insula, activation in the stressed rats was much greater in amplitude and extent. We have reported CRD-evoked activation of the anterior and posterior insula in normal (nonstressed), male rats
Stressed rats also showed significant CRD-evoked activation in the hypothalamus, bed nucleus of stria terminalis, accumbens nucleus and ventral striatum, but deactivation in the cingulate cortex, which were all absent in the sham rats. The hypothalamus is believed to be modulated by PFC, and in turn regulates activity of descending inhibitory and facilitatory pathways through periaqueductal gray and pontomedullary nuclei
The nucleus accumbens and ventral striatum participate in reward responses and positive emotional states. The accumbens nucleus and ventral striatum are also considered part of the emotional motor system
During visceral noxious stimulation, sham treated rats demonstrated a positive SLFC of the PrL cortex with the thalamus, anterior insula and other cortical areas. This is consistent with an integrative role of PrL/PFC in the processing of visceral input. In addition, the negative connectivity of PrL/PFC with limbic areas (amygdala, hypothalamus) and periaqueductal gray in the midbrain is consistent with PFC-limbic-periaqueductal gray inhibitory modulation
Human functional brain imaging has been extensively applied to investigate central processing and modulation of pain, including visceral pain, and to characterize alterations in central pain responses in functional pain disorders, including IBS
In conclusion, chronic stress induced marked alterations in CRD-evoked functional brain activation characterized by