Ò

PAIN 155 (2014) 755–763

www.elsevier.com/locate/pain

Sex differences in connectivity of the subgenual anterior cingulate cortex Gang Wang a,b, Nathalie Erpelding a, Karen D. Davis a,b,c,⇑ a

Division of Brain, Imaging and Behaviour—Systems Neuroscience, Toronto Western Research Institute, Toronto, ON, Canada Institute of Medical Science, University of Toronto, Toronto, ON, Canada c Department of Surgery, University of Toronto, Toronto, ON, Canada b

Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.

a r t i c l e

i n f o

Article history: Received 22 August 2013 Received in revised form 23 December 2013 Accepted 10 January 2014

Keywords: Functional connectivity Tractography SgACC PAG MCC TPJ Hypothalamus Salience Habituation Raphe Antinociception

a b s t r a c t We previously reported that women exhibit greater heat pain adaptation to a prolonged painful stimulus and greater habituation to repeated painful stimuli than men. The neural mechanism underlying this sex difference is unknown. However, Bingel et al. (2007) have shown that pain habituation after 8 days of daily pain testing is associated with an increase in pain-evoked activity of the subgenual anterior cingulate cortex (sgACC), suggesting that pain habituation may be mediated via connectivity between the sgACC and the descending pain antinociceptive system. Therefore, we investigated whether women have stronger functional connectivity (FC) and greater structural connectivity (SC) compared to men between the sgACC and the descending antinociceptive system. Our analyses revealed that 1) women exhibited greater FC between the sgACC and the periaqueductal gray (PAG), raphe nucleus, medial thalamus, and anterior midcingulate cortex (aMCC) than men; 2) men had stronger sgACC FC with the anterior insula and temporoparietal junction than women; 3) women and men exhibited comparable SC of the sgACC with the PAG, thalamus, aMCC, anterior insula, and amygdala; and 4) men have stronger sgACC SC with the hypothalamus than women. These data indicate that brain circuitry in women may provide for greater engagement of the descending modulation system mediating pain habituation. In contrast, in men, the salience network may be more engaged, which could support greater sustained attention to pain, thereby preventing pain habituation. Furthermore, the hypothalamus findings suggest a more powerful stress and endorphin-based system at play in men than women. Ó 2014 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.

1. Introduction Pain perception is determined by a complex interaction between ascending pain pathways transmitting nociceptive signals from the body to the brain and descending modulatory cortical and subcortical circuits. As such, in some situations pain can adapt to sustained noxious stimuli or can habituate to repeated noxious stimuli. This pain attenuation involves endogenous modulation mechanisms [11], including the descending pain modulation pathway, with main hubs in the periaqueductal gray (PAG) and the raphe nucleus that modulates the activity of nociceptive neurons in the dorsal horn through serotonergic projections [9,20]. Dysfunctional pain habituation, possibly due to impaired descending modulation, has been linked to chronic pain [40,77,81,100]. Thus, uncovering mechanisms of pain habituation has clinical implications. ⇑ Corresponding author at: Division of Brain, Imaging and Behaviour—Systems Neuroscience, Toronto Western Research Institute, Toronto Western Hospital, 399 Bathurst St, Room MP14-306, Toronto, ON M5T 2S8, Canada. Tel.: +1 416 603 5662; fax: +1 416 603 5745. E-mail address: [email protected] (K.D. Davis).

A pivotal functional magnetic resonance imaging (fMRI) study has shown that pain habituation over the course of 8 days is associated with an increase in pain-evoked activity of the subgenual anterior cingulate cortex (sgACC) [24,25]. Given the role of the sgACC and the descending modulation network in pain habituation, these findings suggest a possible connection between the sgACC and the descending pain antinociceptive system mediating pain habituation. More recently, Hashmi and Davis reported that women exhibit greater heat pain adaptation to a prolonged painful stimulus and greater habituation to repeated painful stimuli compared to men [45]. The neural mechanism underlying this sex difference in habituation remains unknown, but is potentially mediated through sgACC connectivity with antinociceptive pathways. The sgACC is defined as the part of the anterior cingulate cortex (ACC) below the genu or ‘‘knee’’ of the corpus callosum. The sgACC consists of BA25 as well as the subgenual parts of BA32 and of BA24 [50,102]. The term ‘‘subcallosal cortex’’ is often used interchangeably with sgACC. However, the subcallosal cortex is part of the sgACC [74] adjacent to the posterior parolfactory sulcus.

http://dx.doi.org/10.1016/j.pain.2014.01.005 0304-3959/$36.00 Ó 2014 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.

756

Ò

G. Wang et al. / PAIN 155 (2014) 755–763

Here, we will use the term sgACC because our study aimed to examine this region. The aim of this study was to investigate whether there are sex differences in the functional (FC) and structural connectivity (SC) of the sgACC to other brain regions implicated in pain and its modulation. Using resting-state fMRI and diffusion-weighted imaging (DWI)-based probabilistic tractography, we tested the hypothesis that women have stronger FC and SC than men between the sgACC and brain areas implicated in pain processing such as the thalamus (Th), and regions associated in antinociception, including the PAG, raphe nucleus, insula, amygdala, and hypothalamus. 2. Methods 2.1. Participants MRI data were examined from a cohort of 80 healthy righthanded subjects who provided informed written consent to experimental procedures approved by the University Health Network Research Ethics Board. The subject pool consisted of 40 women and 40 men aged between 19 and 36 years (mean age ± SD = 24.5 ± 4.9 years). Subjects were excluded if they reported any current or regular pain (other than menstrual cramps) in the last 6 months (eg, headache, toothache), pain lasting more than 3 months in the last year, any current or previous diagnosis of a psychiatric disorder (eg, depression, attention deficit hyperactivity disorder), any chronic illness, claustrophobia, braces or metal in their body, possibility of pregnancy, or medication/drug use at the dose, frequency, and duration potentially impacting pain or cognitive function. The analysis of sex differences reported here represents a secondary analysis of a psychophysical and imaging dataset examined previously for pain-attention mechanisms [38]. For that study, subjects underwent a psychophysical session to determine heat and cold pain thresholds and temporal summation of heat pain. Stimuli were delivered to the left volar forearm with a 30 mm  30 mm Peltier thermode (TSA-II NeuroSensory Analyzer, Medoc Ltd) to determine pain thresholds and temporal summation of heat pain. Thermal pain thresholds were measured using the methods of limits (for details see [38]). Temporal summation of heat pain was determined using 10 consecutive 48 °C stimuli delivered at 0.5 Hz (interstimulus interval temperature of 40 °C). For each subject, the first 2 blocks of 10 stimuli temporal summation testing were considered training runs, with the third block used to determine temporal summation based on the percentage change of pain rating (on a 0–100 numerical rating scale for ‘‘no pain’’ to ‘‘most intense pain imaginable’’) from the first to the 10th stimulus. Subjects also completed the pain catastrophizing scale questionnaire (PCS) [96]. 2.2. Brain imaging acquisition All imaging data were obtained on a 3T MRI scanner (GE Medical Systems, Milwaukee, WI, USA) fitted with an 8-channel phased-array head coil. For each subject, we obtained 1) a whole-brain high-resolution scan using a T1-weighted inversion recovery prepped, 3-dimensional fast spoiled gradient echo sequence (flip angle = 15°; echo time [TE] = 3 ms; repetition time [TR] = 7.8 ms; inversion time = 450 ms; field of view [FOV] = 25.6 cm; 256  256 matrix; 180 slices; 1-mm slice thickness); 2) a resting-state fMRI scan using a T2⁄-weighted echo-planar imaging sequence (TE = 30 ms; TR = 2000 ms; FOV = 20 cm; 64  64 matrix; 40 slices; 4-mm slice thickness); and 3) 2 diffusion-weighted scans using 60 noncolinear isotropic directions (b = 1000 s/mm2) and 10 nondiffusion-weighted images (b = 0 s/mm2) (TR = 17,000 ms; TE = 83.3 ms; 96  96 matrix; FOV = 23 cm; 64 slices; 2.4-mm slice thickness). For the resting-state scan, subjects were instructed to not think of anything in particular and to keep their eyes closed.

2.3. Resting-state functional connectivity 2.3.1. Definition of seeds A total of 6 bilateral spherical seeds were defined in sgACC regions that have previously been implicated in pain habituation [24,25]. Spheres have been commonly used as a seed shape in fMRI analyses [30,44,53,62,112]. The seeds were drawn as 3-mm spheres centered at Montreal Neurological Institute (MNI) coordinates: A: [ 5, 25, 10]; B: [5, 25, 10]; C: [ 5, 34, 9]; D: [5, 34, 9]; E: [ 6, 33, 9]; F: [6, 33, 9]; G: [ 5, 34, 4]; H: [5, 34, 4]; I: [ 6, 27, 10]; J: [6, 27, 10]; K [ 6, 30, 9]; N: [6, 30, 9] (see Fig. 1, note also that we did not use the letter ‘‘L’’ to avoid confusion with the ‘‘Left’’ designation in data display). We focused our study on FC between sgACC seeds and brain regions associated with the descending modulation system (ie, PAG, raphe nucleus, hypothalamus, amygdala, insula, and Th), although a whole-brain FC analysis allowed us to detect other brain regions to which the sgACC was functionally connected. 2.3.2. Preprocessing and correlation analysis Because of the anatomical location of the sgACC, there was substantial blood-oxygen-level-dependent (BOLD) signal dropout (BOLD signal intensity below 65% of the mean intensity within nonzero intensity voxels; see Supplemental Fig. 1) within the sgACC region of interest for 10 men and 14 women, so these subjects were excluded from analysis. Thus, the resting state fMRI data analysis cohort was comprised of 30 men and 26 women (18– 37 years old, mean ± SD age = 24.6 ± 5.1). Seed-to-voxel correlational analyses were carried out by the functional connectivity (CONN) toolbox Version 13i (http:// web.mit.edu.myaccess.library.utoronto.ca/swg/software.htm) and SPM8. The preprocessing pipeline of the functional images consisted of 1) motion correction, 2) registration to structural images, 3) spatial normalization to the MNI template, 4) smoothing with a Gaussian kernel of 6 mm, and 5) band-pass filtering of 0.01–0.1 Hz as recommended by the CONN toolbox. After these preprocessing steps, the CompCor strategy [19] was implemented, which extracted signal noise from white matter and cerebrospinal fluid by principal component analysis. The analyses did not include global signal regression to avoid potential false anticorrelations (for discussion of this issue, see [73]). Motion parameters, cerebrospinal fluid, and white matter were included in the model and considered as variables of no interest. 2.3.3. Subject- and group-level statistical analyses A first-level analysis was done using the CONN toolbox to perform spatial statistical analyses in each subject. A general linear model was applied to examine significant BOLD signal correlation with respect to time between each seed and each voxel. The toolbox converted the resulting correlation coefficients to Z scores using Fisher’s Z transformation, for subsequent t tests. A higher-level analysis using the CONN toolbox was performed to examine sex differences. To account for multiple comparisons, results were corrected using Monte Carlo simulations implemented in AlphaSim (http://afni.nimh.nih.gov/afni/doc/manual/ AlphaSim) at 10,000 iterations. AlphaSim confirmed that an image-wide threshold of P < 0.05 required a cluster corrected at P < 0.001 with 32 contiguous voxels for significance. 2.4. Probabilistic tractography 2.4.1. Seeds and targets definition For each subject, we performed 2 probabilistic tractography analyses; each tract required the definition of a seed and a target. In our first analysis (analysis I), probabilistic tractography was performed using clusters of significant sex differences that were

Ò

G. Wang et al. / PAIN 155 (2014) 755–763

757

Fig. 1. sgACC seeds and functional connectivity. (A) A total of 6 bilateral seeds, each 3 mm in radius, were used for the resting-state functional magnetic resonance imaging (rs-fMRI) analysis. L, left; R, right. The X, Y, Z coordinates of the seeds are: A: [ 5, 25, 10]; B: [5, 25, 10]; C: [ 5, 34, 9]; D: [5, 34, 9]; E: [ 6, 33, 9]; F: [6, 33, 9]; G: [ 5, 34, 4]; H: [5, 34, 4]; I: [ 6, 27, 10]; J: [6, 27, 10]; K [ 6, 30, 9]; N: [6, 30, 9]. (B) Summary of the main findings of sex differences in sgACC connectivity. In women, sgACC exhibited greater FC to the MD thalamus, PAG, raphe nucleus, and aMCC than in men. However, there was greater sgACC FC in men to the TPJ and aINS than in women. (C) The sgACC seed locations that exhibited sex differences are indicated. Right sgACC seeds are shown in purple and left seeds are shown in green. aINS, anterior insula; aMCC, anterior midcingulate; FC, functional connectivity; MD, medial dorsal thalamic nucleus; PAG, periaqueductal gray; sgACC, subgenual anterior cingulate cortex; TPJ, temporoparietal junction. Brain outline image adapted from http://www2.le.ac.uk/departments/gradschool/training/events/caferesearch/cafe-brain.

identified in our FC analysis. Specifically, we focused on 3 seeds (A/H/N), which had significant FC with the following targets: bilateral temporoparietal junction (TPJ), anterior midcingulate (aMCC), and PAG, respectively. The clusters were chosen to represent the attention/salience and descending pain modulation systems, and to limit the number of tracts to avoid multiple comparison issues. However, since these seeds were too small to identify a significant number of tracts, we performed a refined probabilistic tractography (analysis II) with larger sgACC seeds to increase the tractography sensitivity. Thus, the left and right sgACC seeds were manually defined from the literature [24,25] (MNI coordinates: [9, 30, 12], [ 6, 30, 9], and [3, 36, 12]). To focus on brain regions that are involved in descending modulation, we defined targets in the bilateral anterior insula (aINS), Th, hypothalamus, and amygdala, as well as the PAG and the raphe nucleus (see Supplemental Fig. 2). 2.4.2. Preprocessing Due to signal dropout (DWI signal intensity below 32% of mean intensity within nonzero intensity voxels) in the sgACC seed regions, data from 1 man and 2 women were excluded from the

DWI analysis. Therefore, the final subject cohort used for the SC analysis was comprised of 39 men and 38 women (18–37 years old, mean ± SD age 24.5 ± 5.0 years). Diffusion data were preprocessed by FMRIB’s (Functional MRI of the Brain) Diffusion Toolbox [18,95] (www.fmrib.ox.ac.uk/fsl), performing eddy current distortion correction using affine registration to a b0 image (ie, first b0 collected), brain extraction using Brain Extraction Tool [94], and tract estimations between each seed and target using probabilistic fibre tracking with 5000 tract-following samples and a curvature threshold of 0.2. A dual-fibre diffusion model [17] was fitted to the diffusion data to approximate the probability distribution of diffusion directions for each voxel. Specifically, tractography was conducted by tracing streamline samples from each seed voxel with the trace paths being constrained by the probability distribution. As such, tract-following samples were sent out from every seed voxel. The number of times for which the streamline samples passed a brain voxel correlates to the SC between the voxel and the seed region of interest. This allowed the mapping of an SC distribution from the seed region of interest to the remaining brain. Finally, FLIRT [48,49] was used to

758

Ò

G. Wang et al. / PAIN 155 (2014) 755–763

create affine transformation matrices between individual anatomical space, individual diffusion space, and standard space. In summary, for each subject, we performed 2 probabilistic tractography analyses. In analysis I, we investigated the SC between the sgACC and brain regions that had exhibited sex differences in our FC analysis. In analysis II, we examined SC between the sgACC and brain regions involved in descending modulation processes. To be included in higher-level analyses, a tract needed to hit one target voxel at minimum twice out of 5000 [54]. 2.4.3. Subject- and group-level analyses The number of successful hits from a seed to a target was averaged, yielding the SC for that tract at the subject level. We calculated the common connectivity (CC) of tracts, which provides information about the percentage of subjects that share the same tract [54]. For the group-level analysis, we only included tracts that were common to at least 50% [54] of subjects. CC was used solely for testing the existence of a tract in all subjects, while the number of stream lines, which reached from a seed to a target, was utilized for group level comparison. Sex differences were analyzed using independent-samples t tests, which were corrected for multiple comparisons (P < 0.0017, Bonferroni corrected). 3. Results 3.1. Pain sensitivity There were no statistically significant sex differences in thermal pain thresholds, temporal summation of heat pain, or pain catastrophizing scores in the subjects included in the brain imaging analyses (see Supplemental Table 1).

connected to were predominantly located in the left hemisphere, as we found that the aMCC (in women), aINS (in men), orbitofrontal (BA47, BA45) (in men), and posterior insula (in both sexes) clusters had left lateralized FC with the sgACC. 3.3. Probabilistic tractography Based on our hypotheses and the sex differences identified in our FC study, analysis I was performed to investigate sex differences between sgACC and the PAG, TPJ, and aMCC. In analysis II, we examined SC sex differences between larger sgACC seeds and brain regions involved in the descending modulation (ie, aINS, Th, amygdala, raphe nucleus, hypothalamus, and PAG). Our first probabilistic tractography analysis (analysis I) revealed that the sgACC was structurally connected to the aMCC and PAG (Supplemental Table 5). Additionally, analysis II confirmed structural connections between the sgACC and the aINS, PAG, hypothalamus, amygdala, and lateral and MD thalamus. Furthermore, the sgACC also showed weak SC with the raphe nucleus (Supplemental Table 5). Probabilistic tractography revealed strong SC (>79% CC) between the sgACC and targets: aMCC, PAG, aINS, Th, hypothalamus, and amygdala (see Fig. 3; tractograms for women and men groups can be found in Supplemental Figs. 3 and 4, respectively). Analysis of sex differences showed that sgACC SC was significantly higher in the ipsilateral hypothalamus in men compared to women (P < 0.0017, Bonferroni corrected) (see Fig. 4). Moreover, SC were stronger in men than women 1) between the right sgACC and regions including the right hypothalamus, the left hypothalamus, and right amygdala; and 2) between the left sgACC and left amygdala, but this was found at an uncorrected P < 0.05, and did not survive correction for multiple comparisons.

3.2. Resting-state functional connectivity 4. Discussion Resting-state FC analysis revealed that the sgACC seeds were functionally connected to multiple cortical and subcortical regions. A group analysis of all subjects revealed that the sgACC had significant FC with the MCC, insula, posterior cingulate cortex, and TPJ (for detailed findings, see Supplemental Table 2). We also noted significant sgACC FC with several other areas that were not specifically of interest with our hypotheses such as the fusiform gyrus, parahippocampal gyrus, middle temporal gyrus (BA21), ventral tegmental area, middle frontal gyrus (BA8), precuneus, superior medial frontal lobe, and inferior temporal pole (see Supplemental Table 1). It should also be noted that findings varied for different seed locations within the sgACC, although many of the main findings were found for multiple sgACC seed locations. Examples of the time-series resting-state activity in 2 representative subjects (one man, one woman) are shown in Fig. 2a. In this example, the sgACC (seed H) had greater FC with the raphe nucleus in the female subject (r = 0.82, P < 0.001) compared to the male subject (r = 0.18, P < 0.05). Group analyses revealed that women had greater FC between the sgACC and the aMCC, raphe nucleus, MD thalamus, and PAG compared to men (see Fig. 2b). Additionally, women also exhibited greater sgACC FC with the cuneus, cerebellum, fusiform gyrus, parahippocampal gyrus, and pontine nuclei, compared to men. In contrast, the men exhibited greater sgACC FC with the aINS, TPJ, pars triangularis (BA 45), and orbital area (BA47) compared to the women (see Fig. 2c). Interestingly, sgACC FC varied with the seed location (see Fig. 1c and Supplemental Tables). In general, seeds that were located in anterior parts of the sgACC had greater FC with the aMCC and raphe nucleus in women, and with the aINS in men, whereas seeds in more posterior sgACC regions exhibited greater FC with the PAG and MD thalamus in women, and with the TPJ in men. With regards to laterality, brain regions the sgACC functionally

This study investigated FC and SC of the sgACC and assessed whether there are FC and SC differences between women and men. We found that 1) the sgACC was functionally connected to the PAG, raphe nucleus, MD thalamus, aMCC, TPJ, and aINS in women and in men, 2) women had stronger FC with nodes of the descending antinociceptive and the affective system (ie, PAG, raphe, MD thalamus, and aMCC) compared to men, whereas men exhibited stronger FC with the TPJ and aINS compared to women, 3) the sgACC was structurally connected to the PAG, amygdala, hypothalamus, MD thalamus, aMCC, and the aINS in women and men, and 4) the sgACC SC with the hypothalamus was greater in men than in women (see Figs. 1 and 3). These findings do not reflect pain catastrophizing, pain thresholds, or temporal summation, as no sex differences were found in this study. However, we previously reported greater pain habituation in women compared to men [45]. Therefore, in line with previous findings of greater pain habituation in women than men [45] and long-term habituation and sgACC activation [25,26], our finding of greater sgACC FC with descending pain modulatory pathways in women provides a framework for understanding more effective pain habituation in women. For example, women may more strongly rely on the classic PAG-mediated descending pathway for modulation, whereas stronger sgACC FC with regions implicated in salience and attention networks in men may reflect sustained attention to pain that could hamper pain habituation. Given the role of the hypothalamus in endorphin-mediated antinociception [39,61,66,67], stronger SC between the sgACC and the hypothalamus suggests greater reliance on the hypothalamus-mediated descending modulation pathway in men than women. This study used both FC and SC approaches to evaluate sex differences that provide a framework to understand how pain is

Ò

G. Wang et al. / PAIN 155 (2014) 755–763

759

Fig. 2. Sex differences in sgACC functional connectivity. (A) Representative individual subject examples of the time series of resting state activity within seed-target pairs. FC between sgACC (seed H) and raphe nucleus is stronger in this female subject (r = 0.82) than the male subject (r = 0.18). (B) Females had strong FC compared to males between sgACC (seed H) and aMCC; sgACC (seed H) and raphe nucleus; sgACC (seed A) and MD; and sgACC (seed N) and PAG (P < 0.05 corrected). (C) Males had stronger FC compared to females between sgACC (seed A) and TPJ, and sgACC (seed E) and aINS (P < 0.05 corrected). aINS, anterior insula; aMCC, anterior midcingulate cortex; F, female; FC, functional connectivity; M, male; MD, medial dorsal thalamic nucleus; PAG, periaqueductal gray; sgACC, subgenual anterior cingulate cortex; TPJ, temporoparietal junction.

modulated differently in men and women. The findings from this study expand upon previous studies that examined connectivity of the sgACC but did not consider sex differences [10,97,111]. For instance, a tractography study in a mixed sample of men and women reported SC between sgACC and regions including hypothalamus, orbitofrontal cortex, and amygdala [10], which was reproduced in this study. Other studies of mixed samples of men and women found FC between sgACC and the orbitofrontal cortex [62,97], the temporal pole and medial prefrontal cortex [111], and the ventromedial prefrontal cortex and posterior insula [97]. Here, we provide novel evidence of sex differences within sgACC FC and SC. Specifically, the sex differences were identified in the descending antinociception system, MD thalamic pathway, the attention/ salience system, and the hypothalamic system. Our resting-state FC analysis revealed that, compared to men, women had stronger FC between the sgACC and cortical regions, including PAG and raphe nucleus. The ventrolateral PAG is responsible for facilitating opioid-mediated analgesia as demonstrated by descending pain-modulation models [14,29,59,70,108,109]. Tracer studies have found that the PAG is closely connected to the raphe nucleus [1,5,28,47,51,106,110]. The PAG is engaged both by ascending projections from the spinal cord and by descending

projections from the cortex. Structural and functional animal studies have repeatedly shown that the raphe nucleus is an important intermediary region between the spinal cord and PAG-mediating analgesia [1,8,12,13,15,16,21,22,42,55,58,78,80,82,86–88,92,93]. Accordingly, our findings suggest that women may be able to more strongly engage the descending pain modulation pathway, involving the PAG and raphe nucleus, than men. The cortical pain processing system has been classically divided into 2 sub-systems: the lateral (SI, SII, and lateral thalamus) and the medial system (including MD thalamus, insula, and ACC). The lateral system is believed to play a larger role in the sensory-discriminative pain dimension, while the medial system is thought to be more important in the affective-motivational pain dimension (for review, see [98]). Our finding that women had stronger FC between the sgACC and MD thalamus suggests that the medial pain system may be more effective in women, thus possibly contributing to their greater pain habituation. Our resting-state FC analysis also identified that men had stronger sgACC connectivity with the aINS and TPJ compared to women. The aINS and TPJ are involved in salience detection [35– 37] and considered to be part of the ventral attention network [32,46,56]. Tracer studies in monkeys indicate that the MCC

760

Ò

G. Wang et al. / PAIN 155 (2014) 755–763

Fig. 3. Group anatomical connectivity between sgACC seeds and targets. The absolute lower threshold for common connectivity was 50%, or 39/77 subjects (threshold increased in some instances for display purposes). Tractograms are projected to the MNI152 standard brain (2 mm, T1) image provided in FSL. Colour flares below images represent the number of subjects. aINS, anterior insula; aMCC, anterior midcingulate; Amy, amygdala; Hy, hypothalamus; L, left; PAG, periaqueductal gray; R, right; sgACC, subgenual anterior cingulate; Th, thalamus.

strongly interconnects with the sgACC, the orbitofrontal cortex, and dysgranular and agranular parts of the insula (for review see [68]). Additionally, lesion studies in monkeys and tractography studies have shown SC between the TPJ and aINS [75,89,90,99]. Given this evidence, our findings suggest that males’ stronger activation of the attention network may be associated with a greater engagement of their attention to pain, thereby reducing pain habituation. Our probabilistic tractography analysis showed that, compared to women, men had stronger SC between the sgACC and the hypothalamus, spanning the entire left sgACC. Animal studies provide support for connectivity between the sgACC and the hypothalamus [23,31,33,52,84]. Interestingly, the hypothalamus is likely involved in antinociception as it contains l-opioid receptors, which bind to endomorphins and produce analgesia [43]. b-endorphin can also bind and activate l-opioid receptors, and its analgesic effect is well known [65,79,85,107]. Animal studies provide additional anatomical evidence that the hypothalamus uses b-endorphin to facilitate pain modulation [39,61,66,67]. Intriguingly, hypothalamic activity is associated with the onset of cluster headaches, a chronic pain disorder predominantly occurring in men [57,63,83,101]. In addition to nociceptive processes, the spinohypothalamic tract has been postulated to participate in autonomic and endocrine regulation, as well as affective pain processing [2,27,34,60,71]. The hypothalamus participates in the sympathomedullary pathway that activates the adrenal medulla, causing

the release of adrenalin and a sympathetic response. Blockage of a-adrenergic receptors also leads to a greater blood pressure drop in men than women, implying stronger dependence of sympathetic nerve activity in men, for example, for blood pressure regulation, than in women [91]. This effect fits our results of stronger SC between the sgACC and hypothalamus and thus, stronger facilitation of the sympathetic response in men compared to women. In summary, the hypothalamus may be involved in pain perception and associated stress and/or sympathetic responses. It should be noted that we found different patterns of connectivity in different subregions of the sgACC. Previous studies also reported sgACC subregions that project to different brain regions and govern various functions. For instance, animal studies provide evidence for the anatomical connection between BA32 (seed N) and PAG, and possible functions of this connection [3,105]. Anterograde and retrograde studies in monkeys have revealed reciprocal axonal interconnections among BA24 (seeds C/E/H), BA25 (seeds A and B), and BA32 [6,7,76,103]. BA25 and BA32 directly project axons to the medial and dorsal raphe nucleus [31,41]. Thus, BA25 and BA32 may act as relay stations between BA24 and the raphe nucleus. In retrograde tracer studies of rhesus monkeys, the magnocellular and parvicellular parts of the MD thalamus, as well as the entorhinal cortex, were found to project to BA25 (seed A) [4,104]. Similarly, BA25 projects to anterior hypothalamus, PAG, amygdala, and MD thalamus [41]. The insula projects to BA24 (seed C/E/H) and BA32 [64,69,72,76,104], and BA24, BA25 (seed B), and BA32

Ò

G. Wang et al. / PAIN 155 (2014) 755–763

761

Taken together, although we did not directly measure pain habituation in this study cohort, our data nonetheless provide a neural basis for speculating about sex differences in pain habituation. First, women’s stronger sgACC FC with the descending pain modulation areas (ie, raphe nucleus, PAG) could contribute to a greater pain modulation efficacy compared to men. Second, greater sgACC FC with the MD thalamus in women may enhance modulation of the affective dimension of pain. Third, stronger sgACC FC in men with the salience/attention network (aINS, TPJ) may heighten and sustain their attention to pain. Fourth, stronger sgACC SC in men with the hypothalamus suggests a greater preference for using the possibly slower hypothalamus-mediated descending modulation pathway than the arcuate-mediated descending modulation pathway in women. It is conceivable that the PAG-nucleus raphe pathway and hypothalamus represent 2 separate pain modulation systems, providing further evidence that sex differences reported here may be related to pain. Overall, our findings suggest that women and men may preferentially rely on different pain modulation systems as described by Lovick [59]. Specifically, women may more heavily use the sgACC-ventrolateral PAG-raphe nucleus network, whereas men may rely more on the hypothalamus-lateral PAG-RVM system. Although the seeds used in FC and SC analyses were in close proximity to each other, they yielded some unique findings. This can be attributed to the fact that these 2 techniques measure related but somewhat different neural processes: SC providing insight into anatomical connections that represent basic scaffolding between brain regions, whereas FC reflects how cortical brain regions functionally interact with one another without necessarily being structurally connected. A future study is therefore needed to link functional and structural connectivity in men and women to their individual capacity for pain habituation and/or adaptation. Conflict of interest The authors report no conflict of interest. Acknowledgements This study was supported by the Canadian Institutes of Health Research. Mr. Wang was funded by Ontario Graduate Scholarship and Institute of Medical Science Entrance Scholarship. Dr. Nathalie Erpelding was supported by the Fonds National de la Recherche Luxembourg (PDR-09-023). The authors would like to thank Dr. Adrian Crawley for statistical insights, and Joshua Cheng for some psychophysical data analyses. Fig. 4. Summary of sgACC structural connectivity in the group of all subjects. (A) sgACC exhibited SC (black and blue lines) to aMCC, Th, aINS, Hy, Amy, and PAG. Males showed greater sgACC-SC than females to Hy (P < 0.05 Bonferroni corrected) (blue line). (B) Left sgACC-Hy anatomical connection in males (blue) and females (orange) thresholded at 95% group common connectivity (35/39 subjects in male group; 34/38 subjects in female group). Colour flares below images represent the number of subjects. Tractograms are projected to the MNI152 standard brain (2 mm, T1) image provided in FSL. (C) Paired t test for anatomical connectivity (⁄P < 0.05 Bonferroni corrected). aINS, anterior insula; aMCC, anterior midcingulate; Amy, amygdala; Hy, hypothalamus; PAG, periaqueductal gray; SC, structural connectivity; sgACC, subgenual anterior cingulate; Th, thalamus. Brain outline image in part A adapted from http://www2.le.ac.uk/departments/gradschool/ training/events/caferesearch/cafe-brain.

are strongly interconnected with the orbitofrontal cortex and the insula [68]. Given these extensive structural connections, our results confirm that the sgACC is functionally connected to a number of cortical areas implicated in descending modulation and attention processes.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.pain.2014.01.005. References [1] Abols IA, Basbaum AI. Afferent connections of the rostral medulla of the cat: a neural substrate for midbrain-medullary interactions in the modulation of pain. J Comp Neurol 1981;201:285–97. [2] Abrahams VC, Hilton SM, Zbrozyna A. Active muscle vasodilatation produced by stimulation of the brain stem: its significance in the defence reaction. J Physiol 1960;154:491–513. [3] An X, Bandler R, Ongur D, Price JL. Prefrontal cortical projections to longitudinal columns in the midbrain periaqueductal gray in macaque monkeys. J Comp Neurol 1998;401:455–79. [4] Bachevalier J, Meunier M, Lu MX, Ungerleider LG. Thalamic and temporal cortex input to medial prefrontal cortex in rhesus monkeys. Exp Brain Res 1997;115:430–44. [5] Bandler R, Keay KA. Columnar organization in the midbrain periaqueductal gray and the integration of emotional expression. Prog Brain Res 1996;107:285–300.

762

Ò

G. Wang et al. / PAIN 155 (2014) 755–763

[6] Barbas H, Ghashghaei H, Dombrowski SM, Rempel-Clower NL. Medial prefrontal cortices are unified by common connections with superior temporal cortices and distinguished by input from memory-related areas in the rhesus monkey. J Comp Neurol 1999;410:343–67. [7] Barbas H, Pandya DN. Architecture and intrinsic connections of the prefrontal cortex in the rhesus monkey. J Comp Neurol 1989;286:353–75. [8] Basbaum AI, Fields HL. The origin of descending pathways in the dorsolateral funiculus of the spinal cord of the cat and rat: further studies on the anatomy of pain modulation. J Comp Neurol 1979;187:513–31. [9] Basbaum AI, Fields HL. Endogenous pain control systems: brainstem spinal pathways and endorphin circuitry. Annu Rev Neurosci 1984;7:309–38. [10] Beckmann M, Johansen-Berg H, Rushworth MF. Connectivity-based parcellation of human cingulate cortex and its relation to functional specialization. J Neurosci 2009;29:1175–90. [11] Beecher HK. Pain in men wounded in battle. Ann Surg 1946;123:96–105. [12] Behbehani MM. Effect of chronic morphine treatment on the interaction between the periaqueductal grey and the nucleus raphe magnus of the rat. Neuropharmacology 1981;20:581–6. [13] Behbehani MM. The role of acetylcholine in the function of the nucleus raphe magnus and in the interaction of this nucleus with the periaqueductal gray. Brain Res 1982;252:299–307. [14] Behbehani MM. Functional characteristics of the midbrain periaqueductal gray. Prog Neurobiol 1995;46:575–605. [15] Behbehani MM, Fields HL. Evidence that an excitatory connection between the periaqueductal gray and nucleus raphe magnus mediates stimulation produced analgesia. Brain Res 1979;170:85–93. [16] Behbehani MM, Pomeroy SL, Mack CE. Interaction between central gray and nucleus raphe magnus: role of norepinephrine. Brain Res Bull 1981;6:361–4. [17] Behrens TE, Berg HJ, Jbabdi S, Rushworth MF, Woolrich MW. Probabilistic diffusion tractography with multiple fibre orientations: what can we gain? Neuroimage 2007;34:144–55. [18] Behrens TE, Woolrich MW, Jenkinson M, Johansen-Berg H, Nunes RG, Clare S, Matthews PM, Brady JM, Smith SM. Characterization and propagation of uncertainty in diffusion-weighted MR imaging. Magn Reson Med 2003;50:1077–88. [19] Behzadi Y, Restom K, Liau J, Liu TT. A component based noise correction method (CompCor) for BOLD and perfusion based fMRI. Neuroimage 2007;37:90–101. [20] Beitz AJ. The sites of origin brain stem neurotensin and serotonin projections to the rodent nucleus raphe magnus. J Neurosci 1982;2:829–42. [21] Beitz AJ, Mullett MA, Weiner LL. The periaqueductal gray projections to the rat spinal trigeminal, raphe magnus, gigantocellular pars alpha and paragigantocellular nuclei arise from separate neurons. Brain Res 1983;288:307–14. [22] Beitz AJ, Shepard RD, Wells WE. The periaqueductal gray-raphe magnus projection contains somatostatin, neurotensin and serotonin but not cholecystokinin. Brain Res 1983;261:132–7. [23] Berk ML, Finkelstein JA. Efferent connections of the lateral hypothalamic area of the rat: an autoradiographic investigation. Brain Res Bull 1982;8:511–26. [24] Bingel U, Herken W, Teutsch S, May A. Habituation to painful stimulation involves the antinociceptive system – a 1-year follow-up of 10 participants. PAINÒ 2008;140:393–4. [25] Bingel U, Schoell E, Herken W, Buchel C, May A. Habituation to painful stimulation involves the antinociceptive system. PAINÒ 2007;131:21–30. [26] Bingel U, Tracey I. Imaging CNS modulation of pain in humans. Physiology (Bethesda) 2008;23:371–80. [27] Burstein R, Falkowsky O, Borsook D, Strassman A. Distinct lateral and medial projections of the spinohypothalamic tract of the rat. J Comp Neurol 1996;373:549–74. [28] Cameron AA, Khan IA, Westlund KN, Willis WD. The efferent projections of the periaqueductal gray in the rat: a Phaseolus vulgaris-leucoagglutinin study. II. Descending projections. J Comp Neurol 1995;351:585–601. [29] Cannon JT, Prieto GJ, Lee A, Liebeskind JC. Evidence for opioid and nonopioid forms of stimulation-produced analgesia in the rat. Brain Res 1982;243:315–21. [30] Chang C, Glover GH. Time-frequency dynamics of resting-state brain connectivity measured with fMRI. Neuroimage 2010;50:81–98. [31] Chiba T, Kayahara T, Nakano K. Efferent projections of infralimbic and prelimbic areas of the medial prefrontal cortex in the Japanese monkey, Macaca fuscata. Brain Res 2001;888:83–101. [32] Corbetta M, Patel G, Shulman GL. The reorienting system of the human brain: from environment to theory of mind. Neuron 2008;58:306–24. [33] Csaki A, Kocsis K, Halasz B, Kiss J. Localization of glutamatergic/aspartatergic neurons projecting to the hypothalamic paraventricular nucleus studied by retrograde transport of [3H]D-aspartate autoradiography. Neuroscience 2000;101:637–55. [34] Dostrovsky JO, Craig AD. Ascending projection systems. In: McMahon SB, Koltzenburg M, editors. Wall and Melzack’s textbook of pain. London: Elsevier Churchill Livingstone; 2006. p. 187–93. [35] Downar J, Crawley AP, Mikulis DJ, Davis KD. A multimodal cortical network for the detection of changes in the sensory environment. Nat Neurosci 2000;3:277–83. [36] Downar J, Crawley AP, Mikulis DJ, Davis KD. The effect of task relevance on the cortical response to changes in visual and auditory stimuli: an eventrelated fMRI study. Neuroimage 2001;14:1256–67.

[37] Downar J, Crawley AP, Mikulis DJ, Davis KD. A cortical network sensitive to stimulus salience in a neutral behavioral context across multiple sensory modalities. J Neurophysiol 2002;87:615–20. [38] Erpelding N, Davis KD. Neural underpinnings of behavioural strategies that prioritize either cognitive task performance or pain. PAINÒ 2013;154:2060–71. [39] Fields HL, Basbaum AI, Heinricher MM. Central nervous system mechanisms of pain modulation. In: Koltzenburg M, Wall PD, Melzack R, editors. Wall & Melzack’s textbook of pain. London: Churchill Livingstone; 1999. p. 309–29. [40] Flor H, Diers M, Birbaumer N. Peripheral and electrocortical responses to painful and non-painful stimulation in chronic pain patients, tension headache patients and healthy controls. Neurosci Lett 2004;361:147–50. [41] Freedman LJ, Insel TR, Smith Y. Subcortical projections of area 25 (subgenual cortex) of the macaque monkey. J Comp Neurol 2000;421:172–88. [42] Gebhart GF, Sandkuhler J, Thalhammer JG, Zimmermann M. Inhibition of spinal nociceptive information by stimulation in midbrain of the cat is blocked by lidocaine microinjected in nucleus raphe magnus and medullary reticular formation. J Neurophysiol 1983;50:1446–59. [43] Goldberg IE, Rossi GC, Letchworth SR, Mathis JP, Ryan-Moro J, Leventhal L, Su W, Emmel D, Bolan EA, Pasternak GW. Pharmacological characterization of endomorphin-1 and endomorphin-2 in mouse brain. J Pharmacol Exp Ther 1998;286:1007–13. [44] Grova C, Makni S, Flandin G, Ciuciu P, Gotman J, Poline JB. Anatomically informed interpolation of fMRI data on the cortical surface. Neuroimage 2006;31:1475–86. [45] Hashmi JA, Davis KD. Women experience greater heat pain adaptation and habituation than men. PAINÒ 2009;145:350–7. [46] He BJ, Snyder AZ, Vincent JL, Epstein A, Shulman GL, Corbetta M. Breakdown of functional connectivity in frontoparietal networks underlies behavioral deficits in spatial neglect. Neuron 2007;53:905–18. [47] Henderson LA, Keay KA, Bandler R. The ventrolateral periaqueductal gray projects to caudal brainstem depressor regions: a functional-anatomical and physiological study. Neuroscience 1998;82:201–21. [48] Jenkinson M, Bannister P, Brady M, Smith S. Improved optimization for the robust and accurate linear registration and motion correction of brain images. Neuroimage 2002;17:825–41. [49] Jenkinson M, Smith S. A global optimisation method for robust affine registration of brain images. Med Image Anal 2001;5:143–56. [50] Johansen-Berg H, Gutman DA, Behrens TE, Matthews PM, Rushworth MF, Katz E, Lozano AM, Mayberg HS. Anatomical connectivity of the subgenual cingulate region targeted with deep brain stimulation for treatmentresistant depression. Cereb Cortex 2008;18:1374–83. [51] Keay KA, Feil K, Gordon BD, Herbert H, Bandler R. Spinal afferents to functionally distinct periaqueductal gray columns in the rat: an anterograde and retrograde tracing study. J Comp Neurol 1997;385:207–29. [52] Kita H, Oomura Y. Reciprocal connections between the lateral hypothalamus and the frontal complex in the rat: electrophysiological and anatomical observations. Brain Res 1981;213:1–16. [53] Kong J, Tu PC, Zyloney C, Su TP. Intrinsic functional connectivity of the periaqueductal gray, a resting fMRI study. Behav Brain Res 2010;211:215–9. [54] Kucyi A, Moayedi M, Weissman-Fogel I, Hodaie M, Davis KD. Hemispheric asymmetry in white matter connectivity of the temporoparietal junction with the insula and prefrontal cortex. PLoS One 2012;7:e35589. [55] Lakos S, Basbaum AI. An ultrastructural study of the projections from the midbrain periaqueductal gray to spinally projecting, serotoninimmunoreactive neurons of the medullary nucleus raphe magnus in the rat. Brain Res 1988;443:383–8. [56] Langner R, Kellermann T, Eickhoff SB, Boers F, Chatterjee A, Willmes K, Sturm W. Staying responsive to the world: modality-specific and -nonspecific contributions to speeded auditory, tactile, and visual stimulus detection. Hum Brain Mapp 2012;33:398–418. [57] Leone M, Franzini A, Broggi G, May A, Bussone G. Long-term follow-up of bilateral hypothalamic stimulation for intractable cluster headache. Brain 2004;127:2259–64. [58] Li YQ, Rao ZR, Shi JW. Collateral projections from the midbrain periaqueductal gray to the nucleus raphe magnus and nucleus accumbens in the rat. A fluorescent retrograde double-labelling study. Neurosci Lett 1990;117:285–8. [59] Lovick TA. Integrated activity of cardiovascular and pain regulatory systems: role in adaptive behavioural responses. Prog Neurobiol 1993;40:631–44. [60] Mancia G, Baccelli G, Zanchetti A. Hemodynamic responses to different emotional stimuli in the cat: patterns and mechanisms. Am J Physiol 1972;223:925–33. [61] Mansour A, Fox CA, Akil H, Watson SJ. Opioid-receptor mRNA expression in the rat CNS: anatomical and functional implications. Trends Neurosci 1995;18:22–9. [62] Margulies DS, Kelly AM, Uddin LQ, Biswal BB, Castellanos FX, Milham MP. Mapping the functional connectivity of anterior cingulate cortex. Neuroimage 2007;37:579–88. [63] May A, Bahra A, Buchel C, Frackowiak RS, Goadsby PJ. PET and MRA findings in cluster headache and MRA in experimental pain. Neurology 2000;55:1328–35. [64] Mesulam MM, Mufson EJ. Insula of the old world monkey. III: Efferent cortical output and comments on function. J Comp Neurol 1982;212:38–52. [65] Millan MH, Millan MJ, Herz A. Depletion of central beta-endorphin blocks midbrain stimulation produced analgesia in the freely-moving rat. Neuroscience 1986;18:641–9.

Ò

G. Wang et al. / PAIN 155 (2014) 755–763 [66] Millan MJ. Multiple opioid systems and pain. PAINÒ 1986;27:303–47. [67] Millan MJ. Descending control of pain. Prog Neurobiol 2002;66:355–474. [68] Morecraft RJ, Tanji J. Cingulofrontal interactions and the cingulate motor areas. In: Vogt BA, editor. Cingulate neurobiology and disease. Oxford, UK: Oxford University Press; 2009. p. 113–44. [69] Morecraft RJ, Van Hoesen GW. Convergence of limbic input to the cingulate motor cortex in the rhesus monkey. Brain Res Bull 1998;45:209–32. [70] Morgan MM, Whitney PK, Gold MS. Immobility and flight associated with antinociception produced by activation of the ventral and lateral/dorsal regions of the rat periaqueductal gray. Brain Res 1998;804:159–66. [71] Morton CR, Duggan AW. Inhibition of spinal nociceptive transmission accompanies cardiovascular changes from stimulation in diencephalic ‘defence’ regions of cats. Behav Brain Res 1986;21:183–8. [72] Mufson EJ, Mesulam MM. Insula of the old world monkey. II: afferent cortical input and comments on the claustrum. J Comp Neurol 1982;212:23–37. [73] Murphy K, Birn RM, Handwerker DA, Jones TB, Bandettini PA. The impact of global signal regression on resting state correlations: are anti-correlated networks introduced? Neuroimage 2009;44:893–905. [74] Niida A, Niida R, Matsuda H, Inada T, Motomura M, Uechi A. Identification of atrophy of the subgenual anterior cingulate cortex, in particular the subcallosal area, as an effective auxiliary means of diagnosis for major depressive disorder. Int J Gen Med 2012;5:667–74. [75] Pandya DN, Kuypers HG. Cortico-cortical connections in the rhesus monkey. Brain Res 1969;13:13–36. [76] Pandya DN, Van Hoesen GW, Mesulam MM. Efferent connections of the cingulate gyrus in the rhesus monkey. Exp Brain Res 1981;42:319–30. [77] Peters ML, Schmidt AJ, Van den Hout MA. Chronic low back pain and the reaction to repeated acute pain stimulation. PAINÒ 1989;39:69–76. [78] Pomeroy SL, Behbehani MM. Physiologic evidence for a projection from periaqueductal gray to nucleus raphe magnus in the rat. Brain Res 1979;176:143–7. [79] Porro CA, Cavazzuti M, Baraldi P, Giuliani D, Panerai AE, Corazza R. CNS pattern of metabolic activity during tonic pain: evidence for modulation by beta-endorphin. Eur J Neurosci 1999;11:874–88. [80] Prieto GJ, Cannon JT, Liebeskind JC. N. raphe magnus lesions disrupt stimulation-produced analgesia from ventral but not dorsal midbrain areas in the rat. Brain Res 1983;261:53–7. [81] Proietti Cecchini A, Sandrini G, Fokin IV, Moglia A, Nappi G. Trigeminofacial reflexes in primary headaches. Cephalalgia 2003;23:33–41. [82] Proudfit HK, Anderson EG. Morphine analgesia: blockade by raphe magnus lesions. Brain Res 1975;98:612–8. [83] Qiu E, Wang Y, Ma L, Tian L, Liu R, Dong Z, Xu X, Zou Z, Yu S. Abnormal brain functional connectivity of the hypothalamus in cluster headaches. PLoS One 2013;8:e57896. [84] Room P, Russchen FT, Groenewegen HJ, Lohman AH. Efferent connections of the prelimbic (area 32) and the infralimbic (area 25) cortices: an anterograde tracing study in the cat. J Comp Neurol 1985;242:40–55. [85] Rubinstein M, Mogil JS, Japon M, Chan EC, Allen RG, Low MJ. Absence of opioid stress-induced analgesia in mice lacking beta-endorphin by site-directed mutagenesis. Proc Natl Acad Sci U S A 1996;93:3995–4000. [86] Sandkuhler J, Gebhart GF. Characterization of inhibition of a spinal nociceptive reflex by stimulation medially and laterally in the midbrain and medulla in the pentobarbital-anesthetized rat. Brain Res 1984;305:67–76. [87] Sandkuhler J, Gebhart GF. Relative contributions of the nucleus raphe magnus and adjacent medullary reticular formation to the inhibition by stimulation in the periaqueductal gray of a spinal nociceptive reflex in the pentobarbitalanesthetized rat. Brain Res 1984;305:77–87. [88] Sandkuhler J, Helmchen C, Fu QG, Zimmermann M. Inhibition of spinal nociceptive neurons by excitation of cell bodies or fibers of passage at various brainstem sites in the cat. Neurosci Lett 1988;93:67–72. [89] Saur D, Kreher BW, Schnell S, Kummerer D, Kellmeyer P, Vry MS, Umarova R, Musso M, Glauche V, Abel S, Huber W, Rijntjes M, Hennig J, Weiller C. Ventral and dorsal pathways for language. Proc Natl Acad Sci U S A 2008;105:18035–40. [90] Schmahmann JD. Fiber pathways of the brain. Oxford, UK: Oxford University Press; 2006.

763

[91] Schmitt JA, Joyner MJ, Charkoudian N, Wallin BG, Hart EC. Sex differences in alpha-adrenergic support of blood pressure. Clin Auton Res 2010;20:271–5. [92] Shah Y, Dostrovsky JO. Electrophysiological evidence for a projection of the periaqueductal gray matter to nucleus raphe magnus in cat and rat. Brain Res 1980;193:534–8. [93] Sim LJ, Joseph SA. Efferent projections of the nucleus raphe magnus. Brain Res Bull 1992;28:679–82. [94] Smith SM. Fast robust automated brain extraction. Hum Brain Mapp 2002;17:143–55. [95] Smith SM, Jenkinson M, Woolrich MW, Beckmann CF, Behrens TE, JohansenBerg H, Bannister PR, De Luca M, Drobnjak I, Flitney DE, Niazy RK, Saunders J, Vickers J, Zhang Y, De Stefano N, Brady JM, Matthews PM. Advances in functional and structural MR image analysis and implementation as FSL. Neuroimage 2004;23:S208–19. [96] Sullivan MJ, Adams H, Sullivan ME. Communicative dimensions of pain catastrophizing: social cueing effects on pain behaviour and coping. PAINÒ 2004;107:220–6. [97] Torta DM, Cauda F. Different functions in the cingulate cortex, a metaanalytic connectivity modeling study. Neuroimage 2011;56:2157–72. [98] Treede RD, Kenshalo DR, Gracely RH, Jones AK. The cortical representation of pain. PAINÒ 1999;79:105–11. [99] Umarova RM, Saur D, Schnell S, Kaller CP, Vry MS, Glauche V, Rijntjes M, Hennig J, Kiselev V, Weiller C. Structural connectivity for visuospatial attention: significance of ventral pathways. Cereb Cortex 2010;20:121–9. [100] Valeriani M, de Tommaso M, Restuccia D, Le Pera D, Guido M, Iannetti GD, Libro G, Truini A, Di Trapani G, Puca F, Tonali P, Cruccu G. Reduced habituation to experimental pain in migraine patients: a CO(2) laser evoked potential study. PAINÒ 2003;105:57–64. [101] Vetrugno R, Pierangeli G, Leone M, Bussone G, Franzini A, Brogli G, D’Angelo R, Cortelli P, Montagna P. Effect on sleep of posterior hypothalamus stimulation in cluster headache. Headache 2007;47:1085–90. [102] Vogt BA. Regions and subregions of the cingulate cortex. In: Vogt BA, editor. Cingulate neurobiology and disease. Oxford, UK: Oxford University Press; 2009. p. 3–30. [103] Vogt BA, Pandya DN. Cingulate cortex of the rhesus monkey: II. Cortical afferents. J Comp Neurol 1987;262:271–89. [104] Vogt BA, Pandya DN, Rosene DL. Cingulate cortex of the rhesus monkey: I. Cytoarchitecture and thalamic afferents. J Comp Neurol 1987;262:256–70. [105] Vogt BA, Vogt LJ. Mu-opioid receptors, placebo map, descending systems, and cingulate-mediated control of vocalization and pain. In: Vogt BA, editor. Cingulate neurobiology and disease. Oxford, UK: Oxford University Press; 2009. p. 339–64. [106] Wiberg M, Westman J, Blomqvist A. Somatosensory projection to the mesencephalon: an anatomical study in the monkey. J Comp Neurol 1987;264:92–117. [107] Wu H, Hung K, Ohsawa M, Mizoguchi H, Tseng LF. Antisera against endogenous opioids increase the nocifensive response to formalin: demonstration of inhibitory beta-endorphinergic control. Eur J Pharmacol 2001;421:39–43. [108] Yaksh TL, Yeung JC, Rudy TA. Systematic examination in the rat of brain sites sensitive to the direct application of morphine: observation of differential effects within the periaqueductal gray. Brain Res 1976;114:83–103. [109] Yeung JC, Yaksh TL, Rudy TA. Concurrent mapping of brain sites for sensitivity to the direct application of morphine and focal electrical stimulation in the production of antinociception in the rat. PAINÒ 1977;4:23–40. [110] Yezierski RP. Spinomesencephalic tract: projections from the lumbosacral spinal cord of the rat, cat, and monkey. J Comp Neurol 1988;267:131–46. [111] Yu C, Zhou Y, Liu Y, Jiang T, Dong H, Zhang Y, Walter M. Functional segregation of the human cingulate cortex is confirmed by functional connectivity based neuroanatomical parcellation. Neuroimage 2011;54:2571–81. [112] Zotev V, Krueger F, Phillips R, Alvarez RP, Simmons WK, Bellgowan P, Drevets WC, Bodurka J. Self-regulation of amygdala activation using real-time FMRI neurofeedback. PLoS One 2011;6:e24522.