* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Brain stem representation of thermal and psychogenic sweating in
Single-unit recording wikipedia , lookup
Subventricular zone wikipedia , lookup
Development of the nervous system wikipedia , lookup
Lateralization of brain function wikipedia , lookup
History of anthropometry wikipedia , lookup
Neural engineering wikipedia , lookup
Optogenetics wikipedia , lookup
Neuroscience and intelligence wikipedia , lookup
Donald O. Hebb wikipedia , lookup
Causes of transsexuality wikipedia , lookup
Time perception wikipedia , lookup
Affective neuroscience wikipedia , lookup
Cognitive neuroscience of music wikipedia , lookup
Blood–brain barrier wikipedia , lookup
Evolution of human intelligence wikipedia , lookup
Clinical neurochemistry wikipedia , lookup
Neuromarketing wikipedia , lookup
Emotional lateralization wikipedia , lookup
Activity-dependent plasticity wikipedia , lookup
Nervous system network models wikipedia , lookup
Embodied language processing wikipedia , lookup
Artificial general intelligence wikipedia , lookup
Neurogenomics wikipedia , lookup
Human multitasking wikipedia , lookup
Neuroinformatics wikipedia , lookup
Neural correlates of consciousness wikipedia , lookup
Sports-related traumatic brain injury wikipedia , lookup
Selfish brain theory wikipedia , lookup
Mind uploading wikipedia , lookup
Human brain wikipedia , lookup
Neuroeconomics wikipedia , lookup
Holonomic brain theory wikipedia , lookup
Neuroesthetics wikipedia , lookup
Brain Rules wikipedia , lookup
Neuroplasticity wikipedia , lookup
Neurolinguistics wikipedia , lookup
Cognitive neuroscience wikipedia , lookup
Aging brain wikipedia , lookup
Brain morphometry wikipedia , lookup
Haemodynamic response wikipedia , lookup
Neurotechnology wikipedia , lookup
Neurophilosophy wikipedia , lookup
Neuropsychology wikipedia , lookup
Neuroanatomy wikipedia , lookup
Neuropsychopharmacology wikipedia , lookup
Functional magnetic resonance imaging wikipedia , lookup
Am J Physiol Regul Integr Comp Physiol 304: R810–R817, 2013. First published March 27, 2013; doi:10.1152/ajpregu.00041.2013. Brain stem representation of thermal and psychogenic sweating in humans Michael J. Farrell,1 David Trevaks,1 Nigel A. S. Taylor,3 and Robin M. McAllen1,2 1 Florey Institute of Neuroscience and Mental Health, University of Melbourne, Parkville, Victoria, Australia; 2Department of Anatomy and Neuroscience, University of Melbourne, Parkville, Victoria, Australia; and 3Centre for Human and Applied Physiology, University of Wollongong, Wollongong, New South Wales, Australia Submitted 23 January 2013; accepted in final form 19 March 2013 HUMAN ECCRINE SWEAT GLANDS respond to both thermal and nonthermal drives (26, 27), and their secretions serve a wide range of secondary functions, such as facilitating tactile and thermal sensitivity, increasing contact friction (grip), and reducing the risk of tissue damage (55). The main nonthermal drive to sweating in resting individuals is mental stress or arousal, referred to here as psychogenic sweating (32). Despite earlier doubts (39, 48), it now seems clear that cholinergic sympathetic nerves are the final motor pathways for both thermal and psychogenic sweating (31) and that both types of sweating occur over the whole body rather than being restricted to certain skin areas (32, 33). It is probable (though not yet formally proven) that the same individual sympathetic neurons mediate these responses via common sudomotor units during both thermal and nonthermal sudomotor drives. The central neural pathways that control sweating remain poorly delineated. One can assume that the central neurons controlling thermal and psychogenic sweating are not identical, although they might share common output pathways to the Address for reprint requests and other correspondence: R. McAllen, Florey Institute, Univ. of Melbourne, Parkville Victoria, Australia (e-mail: rmca @florey.edu.au). R810 sympathetic sudomotor nerves. The prevailing view, based on animal studies, is that thermal sweating depends ultimately on the anterior hypothalamus/preoptic area (18, 34, 50, 51), while psychogenic sweating is believed to be driven from the forebrain (21, 44). Clues to the regions of the human brain involved in psychogenic sweating can be found in imaging studies that have related regional brain activity to sweating, usually measured as electrodermal responses. Two elegant brain imaging studies on psychogenic sweating by Critchley and colleagues showed, for example, that the intensity of sweating responses to both mental arithmetic and physical work (handgrip) was associated most clearly with activation of a region in the dorsal midcingulate cortex (5, 7). Other measures of autonomic arousal were also associated with activations in this region, consistent with the view that cortical neurons in that region may ultimately drive psychogenic sweating, perhaps as part of a more general autonomic activation (7). A separate functional magnetic resonance imaging (fMRI) study by the same group showed that regions in the prefrontal and orbitofrontal cortex, as well as extrastriate visual cortex and cerebellum, were activated just before discrete sweating events, implicating neurons in those regions as potential drivers of psychogenic sweating (6). Fechir and colleagues also used fMRI to identify brain regions activated proportionally with sweating in response to a graded mental task (12). Interestingly, while this study showed proportional activation in “task-related” regions, such as the visual, motor, and premotor cortices, as well as part of the cerebellum, it failed to show proportional activation of the mid-cingulate cortex (12). A different approach—EEG dipole analysis—was used by Homma and colleagues to localize cerebral activation associated with sweating events during cognitive processing (21). These experiments indicated involvement of the inferior frontal gyrus, hippocampus, and amygdala (20, 21). The cerebral regions with activity associated with thermal sweating were investigated as part of a separate study by Fechir and colleagues (13), who mapped brain regional metabolic responses with fluorodeoxyglucose positron emission tomography during whole body warming and cooling. These workers identified a small midline region of the posterior cingulate cortex whose metabolic activity was inversely related to sweat rate, but found no cerebral areas with a positive correlation. That posterior cingulate area could represent a locus of inhibitory control of thermal sweating. Alternatively, it might reflect this region’s known role as part of the “default mode circuit” that is inactivated during arousal (47). For the brain stem, limited data from human pathophysiological studies indicate that the descending fiber tracts that supply drive to the spinal preganglionic sudomotor neurons 0363-6119/13 Copyright © 2013 the American Physiological Society http://www.ajpregu.org Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on June 17, 2017 Farrell MJ, Trevaks D, Taylor NAS, McAllen RM. Brain stem representation of thermal and psychogenic sweating in humans. Am J Physiol Regul Integr Comp Physiol 304: R810 –R817, 2013. First published March 27, 2013; doi:10.1152/ajpregu.00041.2013.—Functional MRI was used to identify regions in the human brain stem activated during thermal and psychogenic sweating. Two groups of healthy participants aged 34.4 ⫾ 10.2 and 35.3 ⫾ 11.8 years (both groups comprising 1 woman and 10 men) were either heated by a water-perfused tube suit or subjected to a Stroop test, while they lay supine with their head in a 3-T MRI scanner. Sweating events were recorded as electrodermal responses (increases in AC conductance) from the palmar surfaces of fingers. Each experimental session consisted of two 7.9-min runs, during which a mean of 7.3 ⫾ 2.1 and 10.2 ⫾ 2.5 irregular sweating events occurred during psychogenic (Stroop test) and thermal sweating, respectively. The electrodermal waveform was used as the regressor in each subject and run to identify brain stem clusters with significantly correlated blood oxygen leveldependent signals in the group mean data. Clusters of significant activation were found with both psychogenic and thermal sweating, but a voxelwise comparison revealed no brain stem cluster whose signal differed significantly between the two conditions. Bilaterally symmetric regions that were activated by both psychogenic and thermal sweating were identified in the rostral lateral midbrain and in the rostral lateral medulla. The latter site, between the facial nuclei and pyramidal tracts, corresponds to a neuron group found to drive sweating in animals. These studies have identified the brain stem regions that are activated with sweating in humans and indicate that common descending pathways may mediate both thermal and psychogenic sweating. PLEASE PROVIDE WORDINGS METHODS Participants. Approval for the study was obtained from the Melbourne Health Human Research Ethics Committee (approval no. 2008.147). Two groups of 11 healthy participants were recruited for the two parts of the study: thermal stimulation and mental task. The thermal stimulation group had a mean age of 34.4 ( ⫾ SD 10.2) years, and the mental task group were 35.3 ( ⫾ SD 11.8) years old. Both groups consisted of 10 males and 1 female, with four males participating in both trials. Physiological recordings. Galvanic skin responses (GSR) were used to record sweating events, and these were measured using a pair of Ag-AgCl electrodes (TSD203 electrodes; Biopac Systems, CA, USA) that were positioned on the palmar surfaces of the participants’ right index and middle fingers. The electrodes were connected via a cable (MECMRI-3 MRI cable; Biopac Systems, Goleta, CA, USA) and filter (MRIRFIF interference filter set; Biopac Systems) to a constant voltage amplifier (GSR100C Galvanic Skin Response Amplifier; Biopac Systems). The signal was digitized at 1 kHz (Power 1401; Cambridge Electronic Design, Cambridge, UK) and recorded to computer (Spike2, ver. 7; Cambridge Electronic Design). Artifacts incurred from magnetic field switching during scanning runs were excised from the GSR recording. The electrodermal signal was recorded as an AC signal (0.5-Hz high pass filter) to identify discrete electrodermal events independently of any shift in mean signal level (28). Signals from the scanner control panel were recorded to computer and used to match the timing of electrodermal events to functional brain stem images. Thermal stimulation. Participants wore a body suit (Med-Eng BCS4 Body Cooling System; Allen Vanguard, Ottawa, ON, Canada), incorporating a network of small-diameter (4 mm) plastic tubes through which temperature-controlled water was circulated. Lower and upper garments extended from ankles to the neck, covering both arms to the wrists. Additionally, a woolen blanket was placed over the suited participants, while they were warmed and scanned. The suit was connected to a water reservoir and a pump located in the scanner operating room via 10 m of 15-mm diameter plastic tubing. The water in the reservoir was adjusted to between 40°C and 50°C and was adjusted once sweating commenced (usually after 20 –30 min) to maintain a low mean rate of sweating events. Mental task. Participants performed a color, word Stroop task during scanning. Visual stimuli were projected onto a screen that was visible to participants via a mirror mounted on the head coil. Two different color words written with congruent or incongruent colored letters were presented sequentially in random order for 2-min blocks, interspersed with 30-s rest intervals, during the 7.9-min functional brain-scanning runs. The task was to count the number of nominated events (e.g., count the words “RED” written with yellow letters) in a 2-min block, but the subjects’ responses to the task were not analyzed and did not form part of the experiment. The sequence of events in thermal and psychogenic scanning runs is illustrated in Fig. 1. Functional brain images. The magnetic resonance images used for the study were acquired with a Siemens 3-T Trio system and 32channel head coil located at the Murdoch Children’s Research Institute (Melbourne, Australia). Functional brain stem images were acquired with blood oxygen level-dependent (BOLD) contrast that measures hemodynamic responses subsequent to neural activity (43). The slices of functional brain stem images (echo planar images) were positioned in an oblique, coronal orientation aligned with the dorsal surface of the medulla. Functional brain stem images had 21 slices of 2.5-mm thickness, and covered the entire brain stem upward from the cervical cord. Slices were divided into a 128 ⫻ 128 matrix with a spatial resolution of 1.88 mm ⫻ 1.88 mm. Functional images were acquired every 1.79 s [time to repetition (TR) ⫽ 1790 ms] with a time to echo (TE) of 30 ms and a flip angle of 70°. Scanning runs lasted 7.9 min and incorporated 265 sequential functional brain stem images. Two scanning runs were acquired from all participants during the thermal stimuli and mental tasks. A third scanning run was acquired from three participants during thermal stimulation to ensure that two runs contained sufficient instances of sweating events. Anatomical images acquired for registration. To allow data from different participants to be grouped, each participant’s brain images were warped to the Montreal Neurological Institute (MNI) standard template at a spatial resolution of 1 ⫻ 1 ⫻ 1 mm (4, 11). Two types of anatomical image were acquired to control the warping. First, an echo planar image (EPI) of the whole brain (whole brain EPI) was obtained with similar parameters to the functional brain stem images (2.5-mm oblique coronial slices aligned with the dorsal surface of medulla, 128 ⫻ 128 matrix, in slice resolution 1.88 mm ⫻ 1.88 mm, TE A B Fig. 1. Experimental timelines for thermal (A) and psychogenic (B) scanning runs. AJP-Regul Integr Comp Physiol • doi:10.1152/ajpregu.00041.2013 • www.ajpregu.org Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on June 17, 2017 follow a course through the lateral parts of the medulla oblongata. This is based on the finding that when those areas are damaged, sweating of any cause is abolished on the ipsilateral side of the head and neck (40). The finding that the deficit does not extend below the neck is attributed to spinal sudomotor pathways crossing the midline sufficiently to compensate lower levels of the sympathetic outflow (41). Where those descending fiber tracts originate and the locations of their cell bodies are unknown, however. The present paper is focused on the brain stem, which has not yet received detailed attention in human imaging studies on sweating. We wanted to know the answers to three principal questions: 1) Where are the brain stem neurons that control sweating? 2) Do they differ between thermal and psychogenic sweating? and 3) Can we identify the human homolog of the rostral medullary cell group that drives sweating in cats (49)? R811 R812 PLEASE PROVIDE WORDINGS statistical parametric maps were warped to a standard template brain, as described above. A mask was applied to the grouped data to exclude voxels outside the brain stem. Statistical parametric maps were averaged between two runs for each individual, and then group averages were calculated for all participants in each condition (psychogenic or thermal sweating), as well as for contrasts between the two conditions. Standard methods were used to calculate Z-scores for each voxel in standard brain stem space, accounting for the confounding factors noted above (1). The average statistical parametric map in each condition was then thresholded to include only voxels where Z ⬎ 2.3. Standard procedures were then used to determine significant clusters of activation by applying a cluster corrected threshold of P ⬍ 0.05 (59). To test whether any brain stem region distinguished between thermal and psychogenic sweating events, voxelwise comparisons of BOLD signals were made between sweating-related activity in the two conditions. The results of these comparisons were then thresholded at a cluster-corrected threshold of P ⬍ 0.05. To characterize the time course of sweating-related BOLD signals in regions of interest, (but only for that purpose), a sweating event was defined as an increase exceeding one SD above the baseline noise of the electrodermal signal, and its peak time was used as a trigger point. Electrodermal and preprocessed BOLD signals were averaged over a time window covering 30 s before and after the trigger time. RESULTS Experimental procedure. Sweating events were measured as abrupt increases in skin conductance and were detected as peaks in the electrodermal trace. Representative electrodermal traces from psychogenic and thermal sweating runs (after down-sampling) are shown in Fig. 2. While the baseline was smooth during psychogenic sweating runs (Fig. 2A), low-level ongoing activity was present between the large peaks during thermal sweating (Fig. 2B). That low ongoing activity, indicating small sweating events, would have made a minor contribution to the regressor for brain stem voxel analysis. Both the thermal stimulus and the mental task were adjusted to keep the frequency of large sweating events low, so as to minimize overlap and to preserve the contrast between those events and A B Fig. 2. Examples of electrodermal recordings that were used to extract sweating event-related activity from the brain stem. These show representative records of 7.9-min scanning periods during the mental task (upper trace) and during heating (lower trace). Previously, scanning artifacts were removed from the raw AC recording, and the signal was down-sampled to once per 1.79 s so as to match the imaging sequence (see MATERIALS AND METHODS). The straight dotted line in each trace denotes one SD greater than the mean. Further details are given in the text. AJP-Regul Integr Comp Physiol • doi:10.1152/ajpregu.00041.2013 • www.ajpregu.org Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on June 17, 2017 ⫽ 30 ms, flip angle ⫽ 70°), but incorporating 59 additional slices (80 in total) and taking 5.5 s to acquire (TR ⫽ 5,500 ms). Secondly, a high-resolution, T1-weighted anatomical image of the whole brain was acquired from each participant (192 ⫻ 0.9 mm sagittal slices, 256 ⫻ 256 matrix, in-slice resolution 0.8 mm ⫻ 0.8 mm, TR ⫽ 1900 ms, TE ⫽ 2.59 ms, flip angle ⫽ 9°). Image registration. The transformation of functional brain stem images from each participant to the MNI standard brain was done in three steps using the Functional Magnetic Resonance Imaging of the Brain’s (FMRIB’s) linear image registration tool (FLIRT) (16, 23, 24). In the first step, the middle image of a scanning run (to which all other images in the scanning run had been realigned) was aligned with that participant’s whole brain EPI acquired in the same scanning session. The second step involved coregistration of the participant’s whole brain EPI to the high-resolution T1-weighted anatomical image. Finally, the participant’s high-resolution T1-weighted anatomical image was warped to the MNI standard brain. Matrices representing the transformations of the three steps were multiplied to produce the global transformation from functional brain stem space to standard MNI space. This transformation matrix was applied to statistical parametric maps to perform higher levels of analyses described below. Analysis: preprocessing. Preparation and statistical analysis of functional brain stem images were performed with the Oxford Centre for FMRIB Software Library [Oxford, UK; FSL version 4.1 (http:// www.fmrib.ox.ac.uk/fsl/)]. Sequential functional brain stem images from single-scanning runs were realigned spatially to the middle image of the run, to correct for any head movement during the scanning run using MCFLIRT (23). Images were spatially smoothed using a Gaussian kernel of 3-mm full width at half maximum. The time series of each scanning run was mean-based intensity normalized (all sequential volumes scaled by the same factor) and high pass filtered to remove low-frequency artefacts (58). Statistical analysis. The first level of statistical analysis was performed on individual scanning runs using general linear modeling, including local autocorrelation correction, as instituted in the FMRI expert analysis tool [FEAT, FMRIB’s improved linear model (FILM)] (58). The regressor of interest for analysis was the simultaneously recorded electrodermal signal after scanning artifacts had been removed, and the signal was down-sampled to correspond to the timing of each functional brain stem image (once per 1.79 s). To adjust the appropriate relative timing, two factors were taken into account. The conduction delay between neural activity in the brain and the peak electrodermal response at the finger has been measured as 5 to 5.5 s (21). Therefore, the electrodermal regressors were translated backward in time by 5.37 s (TR ⫻ 3). Additionally, the hemodynamic response as measured by the BOLD signal lags 4 to 6 s after neural activation: compensation for this delay was factored into the standard fMRI analysis (19). To remove the influence of known confounding factors, other regressors were added to the analysis of individual scanning runs. The brain stem is susceptible to physiological noise, including respiratory effects on local magnetic field properties and changes in the blood and cerebrospinal fluid associated with the cardiac cycle. To reduce the effects of physiological noise, regressors from three chosen regions were used in the model to account for variance associated with the cardiac and respiratory cycles, according to the procedures described by Birn et al. (3). Movement is also prevalent in the brain stem, so in addition to realignment of images as a preprocessing procedure (see above), the effects of movement were taken into account by including the six motion parameters (three translations and three rotations) into the modeling of signal changes. The modeling of variance in BOLD signals across a time series in each scanning run was performed on a voxel-by-voxel basis. A parameter estimate was calculated for the fit of each regressor to the observed BOLD signal for each voxel in the native space of the preprocessed functional brain stem images, making a statistical parametric map for each scanning run. For subsequent analyses, these PLEASE PROVIDE WORDINGS Table 1. Clusters of significant activation during thermal and psychogenic sweating Coordinates Region Psychogenic Dorsal Midbrain Dorsal Midbrain Lateral Pons Rostral Lateral Medulla Rostral Lateral Medulla Caudal Lateral Medulla Thermal Dorsal Midbrain Dorsal Midbrain Ventral Midbrain Ventral Midbrain Lateral Pons Rostral Lateral Medulla Rostral Lateral Medulla Caudal Ventral Medulla Caudal Ventral Medulla Side x y z Z score R L L R L R 2 ⫺5 ⫺9 7 ⫺5 9 ⫺28 ⫺24 ⫺29 ⫺35 ⫺35 ⫺42 ⫺3 ⫺4 ⫺43 ⫺48 ⫺48 ⫺50 3.99 3.81 3.67 3.42 3.19 3.58 R L R R L R L R 0 1 ⫺2 3 7 ⫺6 5 ⫺6 3 ⫺29 ⫺31 ⫺20 ⫺24 ⫺25 ⫺33 ⫺35 ⫺40 ⫺39 ⫺5 ⫺5 ⫺19 ⫺20 ⫺38 ⫺46 ⫺48 ⫺58 ⫺61 4.37 3.70 3.41 3.52 3.53 3.80 3.84 3.33 3.27 The table gives the coordinates of the voxel with the highest Z-score in each cluster and value of that Z-score. The coordinates correspond to the standard brain template developed by the Montreal Neurological Institute (11). The three-dimensional coordinate space has its origin located at the anterior commissure: x values are mm left (⫺) or right (⫹) of the anterior commissure; y values denote millimeters anterior (⫹) or posterior (⫺) to the anterior commissure, z gives millimeters above (⫹) or below (⫺) the anterior commissure. ored green in Fig. 3. Second, on the basis that sweating events occur bilaterally over the body but the descending brain stem tracts drive sweating unilaterally (see introduction), we infer that the medullary pathways on the left and the right sides must be activated together (by an unidentified antecedent source). Therefore, we sought areas that were activated roughly symmetrically on either side of the brain stem. This “logical filter” highlighted two areas: the dorsal midbrain and the rostral lateral medulla. Dorsal midbrain activation. The dorsal midbrain region showing bilateral consensus activation was ventral to the superior colliculi, level with the rostral end of the cerebral aqueduct. The locus of consensus activation appeared to be centered lateral to the periaqueductal gray matter (PAG) but may have involved lateral parts of the PAG. The mean BOLD signals extracted from these voxels during thermal and psychogenic sweating events are shown in Fig. 4, A and B. Their relative timing coincided closely, reflecting the fact that the hemodynamic response to generate the BOLD signal and the conduction-plus-neuroffector time to generate the electrodermal response both imposed delays of about 5 s. The peak increase in mean BOLD signal there was 0.7 to 1.0%. Rostral medullary activation. This bilateral region of consensus activation extended through the rostral medulla up to the lower pons. This cluster was centered near the lateral margin of the brain stem, ventral to the facial nucleus and dorsal to the pyramidal tract. The mean BOLD signals from this region are shown in Fig. 4, C and D. The peak signal increase here was 1.2 to 1.4%. DISCUSSION So far as we are aware, this is the first neuroimaging study to investigate the human brain stem regions involved in sweating, although a number of studies have investigated the forebrain structures involved (e.g., 8, 14, 37, 46, 57). In answer to the first question that we posed, we found several regions in the midbrain, pons, and medulla that were activated in association with thermal and psychogenic sweating (detailed in Table 1). In answer to our second question—Was any region associated specifically with psychogenic or thermal sweating?—no region showed any such preference. To our final question—Could we identify the human homolog of the rostral medullary cell group that drives sweating in cats (49)?—the answer is yes (discussed below). Sweating occurs in bursts (54), driven by the bursting pattern of sympathetic sudomotor nerve activity (2). Each burst generates a volume of sweat, which dilates the sweat duct and lowers transdermal resistance, measured as the electrodermal response. At low levels of sweating (as here) the precursor sweat is reabsorbed and disperses over the next few seconds, the duct collapses, and skin conductance falls again. An ACcoupled skin resistance signal gives an index of bursts of sudomotor nerve activity (28), and we have used it here to measure the strength and timing of sweating bursts. Because sweating bursts show clear synchrony across body regions (17, 38, 45, 56), the electrodermal signal from the fingers of one hand served as a suitable measure of the whole body response. The experimental design used these sporadic sweating events, which were unsynchronized to any experimental intervention, as a regressor; established methods were then used to identify AJP-Regul Integr Comp Physiol • doi:10.1152/ajpregu.00041.2013 • www.ajpregu.org Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on June 17, 2017 the intervening periods. The mean numbers of sweating events during each 7.9-min scanning sequence that exceeded the mean level by 1 SD (horizontal lines in Fig. 2) were 7.3 ⫾ 2.1 and 10.2 ⫾ 2.5 for psychogenic and thermal sweating runs, respectively. Brain stem sites activated in association with thermal and psychogenic sweating events. The brain stem regions that showed significant clusters of voxels activated in association with thermal and psychogenic sweating events are listed in Table 1. The principal regions are illustrated in Fig. 3 and will be described further below. No brain stem clusters were found to show a significant reduction in BOLD signal (“deactivation”) associated with electrodermal events during either thermal or psychogenic sweating. To test whether any brain stem regions were preferentially activated with thermal or with psychogenic sweating events, a voxelwise comparison was made between sweating-related BOLD signal changes in the two conditions. This revealed no brain stem region that was activated significantly more during thermal than psychogenic sweating events, or vice versa (P (corrected) ⬎ 0.05). Principal regions of activation. Although there was no statistical difference between regions activated during thermal and psychogenic sweating events, the thresholded activation maps from each condition were not identical. As with other human fMRI studies, such minor anomalies are to be expected when detecting small signals against a noisy background (25). Two features were used to select the activated regions that were most likely to be significant. First, on the basis that thermal and psychogenic sweating events were evidently equivalent (see above) “consensus” voxels that were activated significantly by both stimuli were identified. These were col- R813 R814 PLEASE PROVIDE WORDINGS Thermal Psychogenic A B L z= -4 x= -5 C L brain stem regions whose BOLD signals were correlated with that regressor. The generator of sweating bursts is unknown. At high rates of sweating, they become regular and occur at ⬃80 –100 bursts/min (2, 42), suggesting that they are driven by a central “oscillator”. While it is conceivable that left-right connections between the bilateral regions identified here could generate that rhythm [by analogy with the respiratory rhythm (52)], it seems more likely that they receive periodic drive from an oscillator located elsewhere. This inference is based on the finding that forebrain structures also show activity related to sweating bursts (6), and the likelihood that the pontomedullary locus of activation represents the descending output pathway from the brain (see below). Further work is needed to answer this. To the best of our knowledge the midbrain site we identified as activated with sweating has not been highlighted as such in previous human work. In cats, Davison and Koss (10) used electrical stimulation to map sudomotor pathways from the hypothalamus to the caudal medulla and found that the excitatory tract passed through the midbrain ventrolaterally to the periaqueductal gray matter. Their use of electrical stimulation prevented them from identifying synaptic relays in the pathway, because this method does not distinguish between cell bodies and fiber tracts. An excitatory BOLD signal, however, suggests a synaptic relay. The anatomical correspondence between the cat and human data is plausible, so this region D y= -35 L z= -48 could represent a synaptic relay in the descending sudomotor pathway. By contrast, the site of strong bilateral activation in the rostral medulla was expected on the basis of animal work. McAllen (35) identified a region near the ventral surface of the rostral medulla, which, when activated with microinjections of excitant amino acid, drove sweating in the cat’s paw. That region was anatomically distinct from, and rostromedial to, the vasomotor premotor neurons of the rostral ventrolateral medulla (RVLM; subretrofacial nucleus) (9). A subsequent study on cats (49) mapped the sudomotor cell group to a discrete locus between the facial nucleus and the pyramidal tract. Excitant amino acids activate neurons but not fiber tracts (15), so this juxtafacial region represents a synaptic relay within the cat sudomotor pathway. The bilateral region with sweatingrelated activity identified here in the human rostral medulla was also centered between the facial nucleus and the pyramidal tract (Fig. 3). The strong sweating-related BOLD signal in the juxtafacial region of the human medulla (Fig. 4) also indicates the presence of a synaptic relay rather than a fiber tract. Limitations. As with all human imaging studies, the conclusions reached here are based on correlation rather than proof of causation. In the case of psychogenic sweating, for example, it is conceivable that each sweating event signals a “miniarousal,” to which the medullary and midbrain regions identified here respond in parallel with, but not directly linked to, AJP-Regul Integr Comp Physiol • doi:10.1152/ajpregu.00041.2013 • www.ajpregu.org Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on June 17, 2017 Fig. 3. Map of sweating event-related activity in the brain stem. Significant clusters of activation during thermal sweating are shown in red, and those during psychogenic sweating are shown in blue. Voxels in which both achieved significance are shown in green. A: sagittal slice 5 mm to the left of the midline (x ⫽ ⫺5) shows the rostrocaudal locations of the two main regions of activation that were common to thermal and psychogenic sweating. The yellow circle in the upper part of the figure encloses a region of common activation in the midbrain and the lower circle indicates a common region in the rostral medulla. B: axial (horizontal) section through the midbrain (z ⫽ ⫺4, corresponding to upper green line in A) shows the bilateral symmetry of the common region of activation. C: coronal slice (y ⫽ ⫺35, corresponding with lower vertical green line in A) shows the left-right symmetry of the common rostral medullary activations immediately caudal to the pons. D: dorsoventral location of those common rostral medullary activations can be seen in this axial section (z ⫽ ⫺48, corresponding to lower horizontal green line in A). Note: both axial sections are displayed dorsal side uppermost. Both R815 PLEASE PROVIDE WORDINGS A B D sweating. Indeed, this could also apply during thermal sweating. However, previous work has demonstrated that transient sweating events and “arousal” (measured by mean skin conductance) are linked to activation of distinct cerebral regions (37). Sweating events can, thus, be used to discriminate sweating-related brain activity over and above any background arousal. This suggests that the BOLD signals correlated with sweating events reflect sweating per se. The participants were unselected volunteers, with only one female in each group. Thus, although we expect brain stem mechanisms to apply equally to both sexes, we could not test this from our sample. We recorded electrodermal responses from the palmar surface of fingers on one hand, and used this as a general index on the basis that sweating events across the body occur synchronously. However, sweating responses may differ in amplitude over body regions (33). Therefore, if any body map exists within brain stem sweating pathways, our view would be biased toward the hand. Perspectives and Significance It is encouraging to find the human homolog of a brain stem cell group identified in animals: it gives confidence that both findings are correct and demonstrates the evolutionary conservation of a basic physiological mechanism. In two previous cases, fMRI studies of the human brain stem have found a remarkably accurate correspondence between functional maps of autonomic control and anatomical predictions based on animal studies. These were the demonstration of the involvement of the rostral medullary raphé in cold-defense (36) and of the RVLM (and caudal ventrolateral medulla) in vasomotor control (29, 30). We consider that the sweating-related juxtafacial activation identified here adds a third example. Animal studies indicate that, like the cell groups in the rostral medullary raphé and RVLM, neurons in this region (called rostral ventromedial medulla, RVMM, in rats) send their axons directly to sympathetic preganglionic neurons (22, 53). This juxtafacial region in the human brain stem may, therefore, contain the final common premotor pathway that transmits to the spinal sympathetic neurons the descending signals that drive both thermal and psychogenic sweating. In each case, it remains to be determined which antecedent pathways drive those neurons to cause sweating. ACKNOWLEGDMENTS We acknowledge the technical expertise provided by Michael Kean of the Children’s Magnetic Resonance Imaging Centre, Melbourne, Australia. GRANTS This work was supported by project Grant 509089, and R. McAllen was supported by Principal Research Fellowship 566667 from the National Health and Medical Research Council. M. J. Farrell received support from the Robert J. Kleberg, Jr. and Helen C. Kleberg Foundation and the G. Harold and Leila Y. Mathers Charitable Foundation. AJP-Regul Integr Comp Physiol • doi:10.1152/ajpregu.00041.2013 • www.ajpregu.org Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on June 17, 2017 C Fig. 4. The average time course of peak blood oxygen level-dependent (BOLD) signal extracted from each side of the midbrain and medullary areas of interest (bilateral green regions in Fig. 3), compared with the corresponding averaged electrodermal responses. R816 PLEASE PROVIDE WORDINGS DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the authors. 21. AUTHOR CONTRIBUTIONS Author contributions: M.J.F., N.A.T., and R.M.M. conception and design of research; M.J.F. and D.T. performed experiments; M.J.F. and D.T. analyzed data; M.J.F., D.T., N.A.T., and R.M.M. interpreted results of experiments; M.J.F. and R.M.M. prepared figures; M.J.F., N.A.T., and R.M.M. drafted manuscript; M.J.F., D.T., N.A.T., and R.M.M. edited and revised manuscript; M.J.F., D.T., N.A.T., and R.M.M. approved final version of manuscript. 22. 23. 24. REFERENCES 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. AJP-Regul Integr Comp Physiol • doi:10.1152/ajpregu.00041.2013 • www.ajpregu.org Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on June 17, 2017 1. Beckmann CF, Jenkinson M, Smith SM. General multilevel linear modeling for group analysis in FMRI. NeuroImage 20: 1052–1063, 2003. 2. Bini G, Hagbarth KE, Hynninen P, Wallin BG. Thermoregulatory and rhythm-generating mechanisms governing the sudomotor and vasoconstrictor outflow in human cutaneous nerves. J Physiol 306: 537–552, 1980. 3. Birn RM, Smith MA, Jones TB, Bandettini PA. The respiration response function: the temporal dynamics of fMRI signal fluctuations related to changes in respiration. NeuroImage 40: 644 –654, 2008. 4. Collins DL, Neelin P, Peters TM, Evans AC. Automatic 3D intersubject registration of MR volumetric data in standardized Talairach space. J Comput Assist Tomogr 18: 192–205, 1994. 5. Critchley HD, Corfield DR, Chandler MP, Mathias CJ, Dolan RJ. Cerebral correlates of autonomic cardiovascular arousal: a functional neuroimaging investigation in humans. J Physiol 523: 259 –270, 2000. 6. Critchley HD, Elliott R, Mathias CJ, Dolan RJ. Neural activity relating to generation and representation of galvanic skin conductance responses: a functional magnetic resonance imaging study. J Neurosci 20: 3033– 3040, 2000. 7. Critchley HD, Mathias CJ, Josephs O, O’Doherty J, Zanini S, Dewar BK, Cipolotti L, Shallice T, Dolan RJ. Human cingulate cortex and autonomic control: converging neuroimaging and clinical evidence. Brain 126: 2139 –2152, 2003. 8. Critchley HD, Melmed RN, Featherstone E, Mathias CJ, Dolan RJ. Volitional control of autonomic arousal: a functional magnetic resonance study. NeuroImage 16: 909 –919, 2002. 9. Dampney RA. The subretrofacial vasomotor nucleus: anatomical, chemical and pharmacological properties and role in cardiovascular regulation. Progr Neurobiol 42: 197–227, 1994. 10. Davison MA, Koss MC. Brainstem loci for activation of electrodermal response in the cat. Am J Physiol 229: 930 –934, 1975. 11. Evans AC, Collins DL, Mills SR, Brown ED, Kelly RL, Peters TM. 3D statistical neuroanatomical models from 305 MRI volumes. Proc IEEE Nucl Sci Symp Med Imag Conf 1813–1817, 1993. 12. Fechir M, Gamer M, Blasius I, Bauermann T, Breimhorst M, Schlindwein P, Schlereth T, Birklein F. Functional imaging of sympathetic activation during mental stress. NeuroImage 50: 847–854, 2010. 13. Fechir M, Klega A, Buchholz HG, Pfeifer N, Balon S, Schlereth T, Geber C, Breimhorst M, Maihofner C, Birklein F, Schreckenberger M. Cortical control of thermoregulatory sympathetic activation. Eur J Neurosci 31: 2101–2111, 2010. 14. Fredrikson M, Furmark T, Olsson MT, Fischer H, Andersson J, Langstrom B. Functional neuroanatomical correlates of electrodermal activity: a positron emission tomographic study. Psychophysiology 35: 179 –185, 1998. 15. Goodchild AK, Dampney RA, Bandler R. A method for evoking physiological responses by stimulation of cell bodies, but not axons of passage, within localized regions of the central nervous system. J Neurosci Methods 6: 351–363, 1982. 16. Greve DN, Fischl B. Accurate and robust brain image alignment using boundary-based registration. NeuroImage 48: 63–72, 2009. 17. Hagbarth KE, Hallin RG, Hongell A, Torebjork HE, Wallin BG. General characteristics of sympathetic activity in human skin nerves. Acta Physiol Scan 84: 164 –176, 1972. 18. Hardy JD, Hellon RF, Sutherland K. Temperature-sensitive neurones in the dog’s hypothalamus. J Physiol 175: 242–253, 1964. 19. Henson RN, Friston KJ. Convolution models for fMRI. In: Statistical Parametric Mapping: The Analysis of Functional Brain Images, edited by Friston K, Ashburner J, Kiebel S, Nichols T, and Penny W. London: Elsevier, 2006, p. 178 –192. 20. Homma S, Matsunami K, Han XY, Deguchi K. Hippocampus in relation to mental sweating response evoked by memory recall and mental calculation: a human electroencephalography study with dipole tracing. Neurosci Lett 305: 1–4, 2001. Homma S, Nakajima Y, Toma S, Ito T, Shibata T. Intracerebral source localization of mental process-related potentials elicited prior to mental sweating response in humans. Neurosci Lett 247: 25–28, 1998. Jansen AS, Wessendorf MW, Loewy AD. Transneuronal labeling of CNS neuropeptide and monoamine neurons after pseudorabies virus injections into the stellate ganglion. Brain Res 683: 1–24, 1995. Jenkinson M, Bannister P, Brady M, Smith S. Improved optimization for the robust and accurate linear registration and motion correction of brain images. NeuroImage 17: 825–841, 2002. Jenkinson M, Smith S. A global optimisation method for robust affine registration of brain images. Med Image Analysis 5: 143–156, 2001. Jezzard P, Clare S. Sources of distortion in functional MRI data. Human Brain Map 8: 80 –85, 1999. Kenny GP, Journeay WS. Human thermoregulation: separating thermal and nonthermal effects on heat loss. Front Biosci 15: 259 –290, 2010. Kondo N, Nishiyasu T, Inoue Y, Koga S. Non-thermal modification of heat-loss responses during exercise in humans. Eur J Appl Physiol 110: 447–458, 2010. Kunimoto M, Kirno K, Elam M, Wallin BG. Neuroeffector characteristics of sweat glands in the human hand activated by regular neural stimuli. J Physiol 442: 391–411, 1991. Macefield VG, Gandevia SC, Henderson LA. Neural sites involved in the sustained increase in muscle sympathetic nerve activity induced by inspiratory capacity apnea: a fMRI study. J Appl Physiol 100: 266 –273, 2006. Macefield VG, Henderson LA. Real-time imaging of the medullary circuitry involved in the generation of spontaneous muscle sympathetic nerve activity in awake subjects. Human Brain Map 31: 539 –549, 2010. Machado-Moreira CA, McLennan PL, Lillioja S, van Dijk W, Caldwell JN, Taylor NAS. The cholinergic blockade of both thermally and non-thermally induced human eccrine sweating. Exp Physiol 97: 930 –942, 2012. Machado-Moreira CA, Taylor NAS. Psychological sweating from glabrous and nonglabrous skin surfaces under thermoneutral conditions. Psychophysiology 49: 369 –374, 2012. Machado-Moreira CA, Taylor NAS. Sudomotor responses from glabrous and non-glabrous skin during cognitive and painful stimulations following passive heating. Acta Physiol (Oxf) 204: 571–581, 2012. Magoun HW, Harrison F, Brobeck JR, Ranson SW. Activation of heat loss mechanisms by local heating of the brain. J Neurophysiol 1: 101–114, 1938. McAllen RM. Action and specificity of ventral medullary vasopressor neurones in the cat. Neuroscience 18: 51–59, 1986. McAllen RM, Farrell M, Johnson JM, Trevaks D, Cole L, McKinley MJ, Jackson G, Denton DA, Egan GF. Human medullary responses to cooling and rewarming the skin: a functional MRI study. Proc Natl Acad Sci USA 103: 809 –813, 2006. Nagai Y, Critchley HD, Featherstone E, Trimble MR, Dolan RJ. Activity in ventromedial prefrontal cortex covaries with sympathetic skin conductance level: a physiological account of a “default mode” of brain function. NeuroImage 22: 243–251, 2004. Nakayama T, Tagaki K. Minute pattern of human perspiration observed by a continuously recording method. Jpn J Physiol 9: 359 –364, 1959. Nakazato Y, Tamura N, Ohkuma A, Yoshimaru K, Shimazu K. Idiopathic pure sudomotor failure: anhidrosis due to deficits in cholinergic transmission. Neurology 63: 1476 –1480, 2004. Nathan PW, Smith MC. The location of descending fibres to sympathetic neurons supplying the eye and sudomotor neurons supplying the head and neck. J Neurol Neurosurg Psychiatry 49: 187–194, 1986. Nathan PW, Smith MC. The location of descending fibres to sympathetic preganglionic vasomotor and sudomotor neurons in man. J Neurol Neurosurg Psychiatry 50: 1253–1262, 1987. Nilsson AL, Nilsson GE, Oberg PA. A note on periodic sweating. Acta Physiol Scan 108: 189 –190, 1980. Ogawa S, Lee TM, Kay AR, Tank DW. Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proc Natl Acad Sci USA 87: 9868 –9872, 1990. Ogawa T. Thermal influence on palmar sweating and mental influence on generalized sweating in man. Jpn J Physiol 25: 525–536, 1975. Ogawa T, Bullard RW. Characteristics of subthreshold sudomotor neural impulses. J Appl Physiol 33: 300 –305, 1972. PLEASE PROVIDE WORDINGS 54. 55. 56. 57. 58. 59. transneuronal pseudorabies viral infections. Brain Res 491: 156 –162, 1989. Takahara K. Observations on the discharge of sweat from a single sweat gland. J Oriental Med 24: 4, 1936. Taylor NAS, Machado-Moreira CA. Regional variations in transepidermal water loss, eccrine sweat gland density, sweat secretion rates and electrolyte composition in resting and exercising humans. Extreme Physiol Med 2: 4, 2013. van Beaumont W, Bullard RW, Banerjee MR. Observations on human sweating by resistance hygrometry. Dermatol Digest: 75–87, 1966. Williams LM, Phillips ML, Brammer MJ, Skerrett D, Lagopoulos J, Rennie C, Bahramali H, Olivieri G, David AS, Peduto A, Gordon E. Arousal dissociates amygdala and hippocampal fear responses: evidence from simultaneous fMRI and skin conductance recording. NeuroImage 14: 1070 –1079, 2001. Woolrich MW, Ripley BD, Brady M, Smith SM. Temporal autocorrelation in univariate linear modeling of FMRI data. NeuroImage 14: 1370 –1386, 2001. Worsley KJ, Evans AC, Marrett S, Neelin P. A three-dimensional statistical analysis for CBF activation studies in human brain. J Cereb Blood Flow Metab 12: 900 –918, 1992. AJP-Regul Integr Comp Physiol • doi:10.1152/ajpregu.00041.2013 • www.ajpregu.org Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on June 17, 2017 46. Patterson JC, 2nd Ungerleider LG, Bandettini PA. Task-independent functional brain activity correlation with skin conductance changes: an fMRI study. NeuroImage 17: 1797–1806, 2002. 47. Raichle ME, MacLeod AM, Snyder AZ, Powers WJ, Gusnard DA, Shulman GL. A default mode of brain function. Proc Natl Acad Sci USA 98: 676 –682, 2001. 48. Robertshaw D. Neuroendocrine control of sweat glands. J Invest Dermatol 69: 121–129, 1977. 49. Shafton AD, McAllen RM. Location of cat brain stem neurons that drive sweating. Am J Physiol Regul Integr Comp Physiol (March 6, 2013). doi:10.1152/ajpregu.00040.2013. 50. Shibasaki M, Crandall CG. Mechanisms and controllers of eccrine sweating in humans. Front Biosci (Schol Ed) 2: 685–696, 2010. 51. Smiles KA, Elizondo RS, Barney CC. Sweating responses during changes of hypothalamic temperature in the rhesus monkey. J Appl Physiol 40: 653–657, 1976. 52. Smith JC, Abdala AP, Rybak IA, Paton JF. Structural and functional architecture of respiratory networks in the mammalian brainstem. Philos Trans R Soc Lond B Biol Sci 364: 2577–2587, 2009. 53. Strack AM, Sawyer WB, Hughes JH, Platt KB, Loewy AD. A general pattern of CNS innervation of the sympathetic outflow demonstrated by R817