Skip to main content
NCBI home page
The Journal of Neuroscience logoLink to The Journal of Neuroscience
. 2024 May 16;44(25):e2342232024. doi: 10.1523/JNEUROSCI.2342-23.2024

Structural Connectivity between Olfactory Tubercle and Ventrolateral Periaqueductal Gray Implicated in Human Feeding Behavior

Guangyu Zhou 1,✉, Gregory Lane 1, Thorsten Kahnt 2, Christina Zelano 1,✉
PMCID: PMC11209663  PMID: 38755004

Abstract

The olfactory tubercle (TUB), also called the tubular striatum, receives direct input from the olfactory bulb and, along with the nucleus accumbens, is one of the two principal components of the ventral striatum. As a key component of the reward system, the ventral striatum is involved in feeding behavior, but the vast majority of research on this structure has focused on the nucleus accumbens, leaving the TUB's role in feeding behavior understudied. Given the importance of olfaction in food seeking and consumption, olfactory input to the striatum should be an important contributor to motivated feeding behavior. Yet the TUB is vastly understudied in humans, with very little understanding of its structural organization and connectivity. In this study, we analyzed macrostructural variations between the TUB and the whole brain and explored the relationship between TUB structural pathways and feeding behavior, using body mass index (BMI) as a proxy in females and males. We identified a unique structural covariance between the TUB and the periaqueductal gray (PAG), which has recently been implicated in the suppression of feeding. We further show that the integrity of the white matter tract between the two regions is negatively correlated with BMI. Our findings highlight a potential role for the TUB–PAG pathway in the regulation of feeding behavior in humans.

Keywords: body mass index, olfaction, olfactory tubercle, periaqueductal gray, reward, structural connectivity

Significance Statement

Increasing evidence suggests that olfaction plays an important role in human feeding behavior. However, the neural underpinnings of this role remain relatively unexplored. Here, we examined the structural connectivity of the olfactory tubercle, which has been implicated in both olfaction and reward, using magnetic resonance imaging. We found that a unique connectivity of the olfactory tubercle with the periaqueductal gray was correlated with body mass index. Our findings highlight a potential role for this pathway in the regulation of human feeding behavior.

Introduction

The olfactory tubercle (TUB), or tubular striatum, receives direct monosynaptic projections from the olfactory bulb (Allison, 1954; Wesson, 2020). Though its role in olfactory processing is not well understood, the TUB, along with the nucleus accumbens, comprises the ventral striatum, an important brain region within reward circuitry, and thus likely plays some role in olfactory-related reward processing (Ikemoto, 2007). For example, rodent work has established that the TUB encodes odor valence (Gadziola et al., 2015; Millman and Murthy, 2020) and processes attraction induced by unlearned odorants (Midroit et al., 2021). However, a growing body of work suggests that the TUB is responsive to nonolfactory stimuli as well and indicates a strong multisensory component—and specifically a nonolfactory component—to neural responses in this region (Ikemoto, 2003; Wesson and Wilson, 2010). Thus, the role of the TUB may be related to striatal functions extending beyond olfaction, but the specifics and behavioral relevance of this nonolfactory significance remain poorly understood.

The bulk of our understanding of the TUB comes from rodent work, with very few human studies focusing on this brain region. In line with rodent work, human TUB is preferentially responsive to attractive odors (Midroit et al., 2021), suggesting that rodent TUB findings may be relatable to humans. However, the human TUB has remained relatively uncharacterized, both in terms of its structure and function. One intriguing and relatively underexplored finding from animal work is the identification of anatomical projections between the TUB and the periaqueductal gray (PAG; Omelchenko and Sesack, 2010; Tripathi et al., 2013; Zhang et al., 2017), a key structure in the modulation and propagation of pain, sympathetic activity, and aversive behaviors (Linnman et al., 2012; George et al., 2019). The structural connectivity between the TUB and PAG and its functional role in olfaction remain poorly understood.

Here we explore the idea that connectivity between the TUB and PAG may relate to feeding behavior in humans. Eating is inherently multimodal (Verhagen and Engelen, 2006; Leclercq and Blancher, 2012), and olfactory circuits are inextricably involved in feeding behavior, as perception of food and neural responses to food odors change with motivational state, such as hunger, satiety, or sleep deprivation (O’Doherty et al., 2000; Howard and Gottfried, 2014; Rolls, 2015; Bhutani et al., 2019; Shanahan et al., 2021; Zhao et al., 2023). Intriguingly, both the TUB and PAG have been implicated in homeostatic control over odor-guided eating behaviors in both animals (Jenck et al., 1987; Hao et al., 2019; Murata, 2020; Shin et al., 2023) and humans (Nolan-Poupart et al., 2013). While the TUB may facilitate eating through hedonic and motivational responses to food-related odors and flavors (Murata et al., 2015; Nogi et al., 2020), the PAG may suppress feeding through activation of local GABAergic cells in the ventrolateral column (Hao et al., 2019). Based on this evidence, we speculate that pathways from the PAG to the TUB may regulate motivational responses to food-related odors and hence that the structural connection between TUB and PAG might relate to eating behaviors in humans.

In this paper, we aimed to accomplish three main goals: first, to explore the structural connectivity between TUB and PAG in humans; second, to look for a medial/lateral organization of this connectivity within the TUB, similar to that reported in rodents; and third, to test the hypothesis that the diffusion tensor imaging (DTI)-based integrity of TUB–PAG tracts correlate with body mass index (BMI), which would suggest that this pathway may be involved in regulation of eating behavior. To pursue these goals, we used structural covariance measures, which reflect the similarity of the macrostructural variations across brain regions, and DTI. Where possible, analyses were performed in two separate datasets, including a large dataset from the Human Connectome Project (Dataset 1; Van Essen et al., 2013) and a smaller dataset collected in-house, designed to maximize signal in human olfactory areas (Dataset 2; Zhou et al., 2019).

Materials and Methods

Ethics statement

For Dataset 1, see the Human Connectome Project ethics statement (Van Essen et al., 2013). For Dataset 2, the study was approved by the Northwestern Institutional Review Board, and voluntary written informed consent was obtained from each participant. The study was conducted according to the principles of the Declaration of Helsinki and the Belmont Report.

Experimental design

Our study contains two independent datasets. For the first, Dataset 1, we selected 336 unrelated participants (181 female, 155 male; age range from 22 to 35 years old) from the Human Connectome Project S900 release (Van Essen et al., 2013). Anatomical T1-weighted image and DTI were available for each participant. Two high-resolution (0.7 mm isotropic voxels) T1-weighted anatomical scans were acquired with identical parameters: repetition time (TR), 2,400 ms; echo time (TE), 2.14 ms; inversion time, 1,000 ms; flip angle, 8°; and field of view (FOV), 224 × 224 mm2. The T1-weighted images were averaged across runs after being aligned with each other (Glasser et al., 2013) using FSL's (FMRIB Software Library; RRID:SCR_002823) flirt (degree of freedom of 6) (Jenkinson et al., 2012). DTI data were acquired (1.25 mm isotropic voxels; TR, 5,520 ms; TE, 89.5; flip angle, 78°; FOV, 210 × 180 mm2; multiband factor of 3) through six runs representing three different gradient tables (b = 1,000, 2,000, and 3,000 s/mm2), with each table acquired once with right-to-left and left-to-right encoding polarities, respectively. Each gradient table includes ∼90 diffusion weighting directions plus 6 b = 0 acquisitions. The imaging data were acquired on a custom 3 T Siemens Skyra scanner, and more details of the data acquisition are available at https://www.humanconnectome.org/study/hcp-young-adult/document/1200-subjects-data-release. The imaging data were minimally preprocessed (Glasser et al., 2013). In brief, the T1-weighted images were corrected for bias field and distortions. For DTI data, the b0 images were intensity normalized across runs; EPI susceptibility, eddy-current–induced distortion, gradient nonlinearities, and subject motion were corrected. The fitting of diffusion tensors and probabilistic diffusion models on corrected data were performed using FSL's dtifit and bedpostx tools, respectively.

In a second dataset (Dataset 2), we included all 25 participants (14 female, 11 male; age range from 19 to 50 years old) from a previous resting functional magnetic resonance imaging study in our lab (Zhou et al., 2019). The T1-weighted images were acquired on a 3 T Siemens scanner using a 3D MPRAGE sequence (1 mm isotropic voxels; TR, 2,300 ms; TE, 2.94 ms; flip angle, 9°; FOV, 256 × 256 mm2; 176 sagittal slices). The T1-weighted images were corrected for bias field using FSL.

Statistical analysis

Structural covariance analysis

To construct structural covariance networks, we calculated the gray matter density using a voxel-based morphometry method that was carried out using FSL's fslvbm (Douaud et al., 2007; http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FSLVBM), an optimized voxel-based morphometry protocol (Good et al., 2001). First, structural images were brain-extracted and gray matter segmented before being registered to the Montreal Neurological Institute (MNI) 152 standard space using nonlinear registration (Andersson et al., 2007). The resulting images were averaged and flipped along the x-axis to create a left–right symmetric, study-specific gray matter template. Second, all native gray matter images were nonlinearly registered to this study-specific template and modulated to correct for local expansion (or contraction) due to the nonlinear component of the spatial transformation. The modulated gray matter images were then smoothed with an isotropic Gaussian kernel with a sigma of 3 mm. We then extracted the average gray matter density for each region of interest (ROI), including the anterior olfactory nucleus, TUB, frontal piriform cortex, and temporal piriform cortex for each participant. In our analysis, the gray matter density was averaged across the left and right hemispheres. The correlation of the gray matter density between the ROI and each voxel of the rest of the brain was calculated using FSL's permutation-based method (randomise). Age, gender, and total intracranial volume, which was estimated using FreeSurfer software (RRID: SCR_001847; Buckner et al., 2004), were included as covariates. Multiple-comparisons correction was performed using threshold-free cluster enhancement method (TFCE; Smith and Nichols, 2009).

Partial correlation analysis of the structural covariance

To test whether the PAG structural covariance was specific to TUB, but not other ventral striatal regions, we performed a partial correlation analysis between TUB and PAG while controlling for the nucleus accumbens. The average gray matter density was extracted for each of the ROIs, including the TUB, nucleus accumbens, and PAG. The nucleus accumbens ROI was derived from Harvard–Oxford subcortical structural atlas (HarvardOxford-sub-maxprob-thr25-2mm), which was available in FSL toolbox. To calculate the partial correlation between the TUB and PAG, confounding factors, including age, gender, total intracranial volume, and gray matter density of the nucleus accumbens, were regressed out using a multiple linear regression method (MATLAB's regress; MathWorks; RRID:SCR_001622) before correlation analysis using Pearson’s correlation coefficient (MATLB's corr). Similarly, age, gender, total intracranial volume, and gray matter density of the TUB were regressed out from both the nucleus accumbens and PAG before the calculation of the correlation coefficient between these two regions.

Gradient of structural covariance analysis

To find the principal modes of the spatial variation in covariance pattern across all voxels in the TUB, we used the diffusion embedding method (Coifman et al., 2005; Kharabian Masouleh et al., 2020). We first calculated the structural covariance between each voxel of the TUB and each voxel of the rest of the brain. A whole-brain mask was created by thresholding the tissue prior image provided by FSL (avg152T1_gray) at a threshold of 100. The connectivity matrix was thresholded at 90% for each voxel, and the resulting matrix was transformed into a symmetric affinity matrix based on cosine similarities of the thresholded connectivity pattern between each pair of voxels. Finally, a low-dimensional embedding representation of the high-dimensional connectivity data was calculated using diffusion map embedding. We set the a parameter for diffusion map embedding to 0.5, which has been suggested to be suitable for brain connectivity data (Kharabian Masouleh et al., 2020). Voxels that have similar connectivity patterns will have similar embedding values. The number of components was set to 3, which was limited by the number of voxels with the TUB. The embedding values were z-score normalized and assigned to each voxel. The gradient analysis was performed using functions adapted from the BrainSpace toolbox (Vos de Wael et al., 2020).

DTI connectivity-based classification

We used FSL's FDT tool to perform DTI connectivity-based classification (Behrens et al., 2003; Johansen-Berg et al., 2004). The PAG was used as the seed mask, and the anterior olfactory nucleus, TUB, frontal piriform cortex, and temporal piriform cortex were used as target masks. We also included the cerebellum as exclusion mask. For each voxel within the seed region, 5,000 streamlines were generated (curvature threshold, 0.2; step length, 0.5; steps per sample, 2,000). To calculate the connectivity of each target to the seed, the resulting volume for each target, which contains the number of streamlines that reach that target, was thresholded at 1. To create a group-level connectivity map to each target, the total number of streamlines for each target was calculated as the sum across all nonzero voxels, which was normalized by the total number of streamlines (waytotal). Then, individual volumes, which were normalized by the total number of streamlines, were transformed to MNI space and averaged across all participants, and the biggest target of the PAG was determined. The connectivity probability of each primary olfactory subregion was calculated as the number of streamlines that reach each target as a proportion of the total number of streamlines (waytotal) seeded from PAG. One-way repeated analysis of covariance (ANOVA) was used compare the connectivity probability across targets, and post hoc analysis was performed using a two-tailed paired t test with multiple comparisons corrected using Bonferroni’s method.

Fiber tractography

To reconstruct the pathways between the TUB and PAG, probabilistic tractography was performed using FSL's FDT tools (Behrens et al., 2003; Johansen-Berg et al., 2004). The TUB was used as the seed region, and PAG was set as the waypoint mask. All the other primary olfactory area regions, including the anterior olfactory nucleus, frontal piriform cortex, and temporal piriform cortex, were included as exclusion masks, as was the cerebellum. For each voxel within the seed region, 5,000 streamlines were generated (curvature threshold, 0.2; step length, 0.5; steps per sample, 2,000). The resulting pathways were thresholded at 1% of the total number of pathways (waytotal) and binarized for each participant. A group-level pathway was generated by averaging individual binarized pathways. Then, second group-level pathways were generated using the PAG as seed region and the TUB as waypoint, while keeping all other parameters the same. Finally, the pathways between the TUB and PAG were calculated as the average of the two group-level pathways.

To reconstruct the pathways of the PAG, we performed probabilistic tractography using the PAG as seed region and whole-brain gray matter as waypoint mask. The whole-brain gray matter mask was created by thresholding the prior gray matter tissue (FSL's avg152T1_gray) at 100. Furthermore, we included olfactory areas as exclusion masks, including the anterior olfactory nucleus, frontal and temporal piriform cortex, TUB, medial amygdala, cortical amygdala, and the periamygdaloid complex (Noto et al., 2021). The parameters of the tractography are as described above. The resulting pathways were thresholded at 1% of the total number of pathways (waytotal) and binarized for each participant. A group-level pathway was generated by averaging individual binarized pathways.

Correlation between BMI and fractional anisotropy

To calculate the group-level correlation between BMI and pathways of the PAG, we averaged the fractional anisotropy of the pathway (thresholded at 25% of total participants) and correlated it with BMI using Pearson’s correlation coefficient (MATLAB's corr). To calculate the voxelwise correlation between BMI and fractional anisotropy, we first smoothed individual fractional anisotropy images using an isotropic Gaussian filter with a sigma of 2 mm. Then we calculated Pearson's coefficient between the BMI and the fractional anisotropy of each voxel of the pathways. Multiple-comparisons correction was performed using the false discovery rate (Benjamini and Hochberg, 1995) method. Age and gender were regressed out using linear regression analysis before the correlation analysis. In the correlation analysis for females and males, only age was regressed out. To compare correlation coefficients between females and males, a statistic z was calculated as z=(z1−z2)/√(1n1−3+1n2−3) where z1 and z2 are the Fisher's transformation of the correlation coefficients and n1 and n2 are the number of samples for each group separately. One participant was excluded from the analysis because the BMI data were not available.

To exclude the effect of whole-brain white matter on the correlation between body mass index and fractional anisotropy of the TUB–PAG pathway, we included the whole-brain fractional anisotropy as a covariate in the correlation analysis. The white matter mask was created by binarizing FSL's FMRIB58_FA-keleton_1mm image, and the average fractional anisotropy of the mask was extracted.

Data availability

The source data related to this study are available at https://github.com/zelanolab/human_olfactory_tubercle_and_periaqueductal_gray_structural_connectivity.

Results

Structural covariance between the olfactory tubercle and PAG

To examine the structural covariance between each subregion of the primary olfactory areas and the rest of the brain, we used voxel-based morphometry. We defined each subregion, including the TUB, anterior olfactory nucleus, and frontal and temporal piriform cortex, of the primary olfactory areas (Mai et al., 2015; Zhou et al., 2019) as ROIs (Fig. 1A). We then calculated the correlation coefficient between the gray matter density of each ROI and every voxel of the whole brain using a permutation method (see Materials and Methods). The results showed a unique, statistically significant (TFCE corrected p < 0.005), structural covariance between the TUB and PAG (Dataset 1; Fig. 1B, left, top row). No significant structural covariance was found between the PAG and other primary olfactory subregions, including the anterior nucleus, frontal piriform cortex, and temporal piriform cortex (TFCE corrected p > 0.005; Dataset 1; Fig. 1B, left, bottom three rows). Identical analyses performed in a second dataset replicated these findings (Dataset 2; Fig. 1B, right). Since TUB is also part of the ventral striatum, along with the nucleus accumbens, we also examined the structural covariance between the nucleus accumbens and the rest of the brain. We found no significant structural covariance between PAG and nucleus accumbens in Dataset 1, but we did observe a small cluster overlapping with PAG in Dataset 2. To test whether the structural covariance between the TUB and PAG could be accounted for by other nearby brain regions of the ventral striatum, we computed a partial correlation of the gray matter density between the TUB and PAG while regressing out that of the nucleus accumbens (Fig. 2A). The structural covariance between the TUB and PAG remained statistically significant although small when accounting for nucleus accumbens gray matter density in both datasets (Dataset 1: Pearson's r = 0.19; p = 5.7 × 10−4; Dataset 2: r = 0.54; p = 0.0052; Fig. 2B). In contrast, the structural covariance between nucleus accumbens and PAG gray matter density was not significant when regressing out TUB gray matter density (Dataset 1: Pearson's r = −0.035; p = 0.52; Dataset 2: r = 0.083; p = 0.69; Fig. 2C). These results suggest specificity of structural covariance between the TUB and PAG relative to other nearby primary olfactory and ventral striatal areas.

Figure 1.

Figure 1.

Open in a new tab

Structural covariance of the primary olfactory areas. A, Seed regions were adapted from Zhou et al., (2019). B, Whole-brain structural connectivity maps of each subregion of the primary olfactory area. Multiple-comparisons correction was performed using threshold cluster-free enhancement (corrected p < 0.005). TUB, olfactory tubercle; AON, anterior olfactory nucleus; PirF, frontal piriform cortex; PirT, temporal piriform cortex; PAG, periaqueductal gray; R, right.

Figure 2.

Figure 2.

Open in a new tab

Structural covariance between the ventral striatum and PAG. A, NAcc region of interest (purple). B, Scatterplot of correlation of the GMD between the TUB and PAG. Age, gender, total intracranial volume, and the GMD of the NAcc were regressed out. C, Scatterplot of the correlation of the GMD between the NAcc and PAG. Age, gender, total intracranial volume, and the GMD of the TUB were regressed out. Each dot in panels B and C represents one participant and the straight line indicates fit line. The correlation was examined using Pearson’s correlation coefficient. PAG, periaqueductal gray; NAcc, Nucleus accumbens; GMD, gray matter density; TUB, olfactory tubercle; L, left.

Lateral-medial organization of the structural covariance of the TUB

Rodent work suggests a medial-lateral structural and functional organization within the TUB (Schwob and Price, 1984; Cansler et al., 2020). To characterize the spatial distribution of the structural covariance with PAG within the TUB, we created a PAG ROI by thresholding the whole-brain structural covariance map at corrected p < 0.005, which was further masked with a PAG atlas mask (Keuken and Forstmann, 2015; Fig. 3A,E). Then, the structural covariance of the PAG ROI with each voxel of the TUB was calculated using Pearson's correlation coefficient. We found that the medial TUB had stronger structural covariance with the PAG than the lateral part (Fig. 3B,D,F,H). This finding was further evidenced by a significant correlation (all p values < 0.005) between the x coordinate of each voxel of the TUB and its structural covariance with the PAG (Fig. 3C,G). To further confirm the lateral-medial organizational pattern of the structural covariance within the TUB, we performed a gradient analysis of the structural covariance using a diffusion embedding method. The results revealed a lateral-medial organization of the first gradient for both Dataset 1 (Fig. 4A) and Dataset 2 (Fig. 4B). The robustness of the gradient analysis was indicated by a statistically significant correlation of the gradients between the two independent datasets (Fig. 4C–E), suggesting an overall similar macrostructural organization of the TUB in humans and rodents.

Figure 3.

Figure 3.

Open in a new tab

Lateral-medial organization of the structural covariance of the TUB with PAG. A, E, Seed region of the PAG created by thresholding the whole-brain structural covariance map of the TUB at corrected p < 0.005, with the resulting image further masked with a PAG mask (pink line) derived from Keuken and Forstmann (2015); thresholded at 25%. B, F, Structural covariance between the PAG and each voxel of the TUB. C, G, Scatterplot of the correlation between the MNI x coordinates of each voxel of the left (n = 23) and right TUB (n = 24) and its structural covariance with the PAG. Each dot represents one voxel and the black line indicates fit line at total. The correlation was examined using Pearson’s correlation coefficient. D, H, 3D plot of the structural covariance between the PAG and each voxel of the TUB shown in B and E separately. Results of Dataset 1 are shown in panels A–D and Dataset 2 in panels E–H. TUB, olfactory tubercle; PAG, periaqueductal gray; R, right; L, left; A, anterior; P, posterior; S, superior; I, inferior.

Figure 4.

Figure 4.

Open in a new tab

Gradient of the TUB whole-brain structural covariance network. A, First gradient value of Dataset 1 overlayed on MNI standard brain (top) and its 3D plot (bottom). B, First gradient value of Dataset 2 overlayed on MNI standard brain and its 3D plot. C, Scatterplot of the correlation of the first gradient values between Dataset 1 and Dataset 2. Each dot represents one voxel. The blue and red lines indicate fit lines for left (n = 23) and right (n = 24) TUB, respectively. The correlation was examined using Pearson’s correlation coefficient. D, Explained variance for each component. E, Scatterplot of the three gradients of the left and right TUB for each dataset. TUB, olfactory tubercle; MNI, Montreal Neurological Institute; R, right; L, left; A, anterior; P, posterior; S, superior; I, inferior.

Fiber streamlines between primary olfactory cortex and PAG are maximal in the TUB and in the ventrolateral column of the PAG

To complement the structural covariance findings between TUB and PAG, we next examined the connection probability between the PAG and subregions of the primary olfactory areas using DTI. We performed a connectivity-based seed classification using the PAG as a seed region, and the subregions of the primary olfactory areas as target regions. The number of fiber streamlines that reached each target ROI was extracted and compared across the target regions. We found that the TUB was the largest target of the PAG (Fig. 5A) and showed statistically significantly stronger probability connectivity than other primary olfactory areas, as revealed by one-way repeated ANOVA (F(3,1005) = 466.55; p = 0; eta2 = 0.58) followed by two-tailed paired t tests (all p values <0.005; Bonferroni corrected; Fig. 5B). The PAG is composed of well-established subregions, which perform different functions. To examine the strength of TUB–PAG streamlines across the PAG, we calculated the spatial distribution of the number of streamlines that reach the TUB target as a proportion of the total number of streamlines (waytotal) seeded from PAG and found that the connection of the PAG with the TUB was maximized in the ventral-lateral column (Fig. 5C).

Figure 5.

Figure 5.

Open in a new tab

DTI connectivity-based classification of the PAG. A, Target regions that receive the largest number of streamlines from the PAG. B, The number of streamlines that reach each primary olfactory subregion as a proportion of the total number of streamlines (waytotal) seeded from PAG. The connectivity probability was different across all targets (two-tailed paired t test; Bonferroni corrected p < 0.005). C, Spatial distribution of the number of streamlines that reach the TUB target as a proportion of the total number of streamlines (waytotal) seeded from PAG. DTI, diffusion tensor imaging; PAG, periaqueductal gray; AON, anterior olfactory nucleus; PirF, frontal piriform cortex; PirT, temporal piriform cortex; R, right.

Probability tractography between the TUB and PAG

To reconstruct the pathways between the TUB and PAG, we performed probabilistic tractography using DTI from Dataset 1. We used the TUB as the seed region and the PAG as a waypoint for fiber tracking and vice versa. The anterior olfactory nucleus, frontal piriform cortex, temporal piriform cortex, and cerebellum were used as exclusion masks. The resulting group-level pathways were then averaged. The analysis revealed a pathway between the TUB and PAG through the medial thalamus (Fig. 6A,B), which has been suggested by both human and rodent studies (Vertes et al., 2015). The group-level pathway between the PAG and whole-brain gray matter, with olfactory areas including the anterior olfactory nucleus, frontal and temporal piriform cortex, TUB, medial amygdala, cortical amygdala, and the periamygdaloid complex as exclusion masks, is shown in Figure 6C.

Figure 6.

Figure 6.

Open in a new tab

Tractography between the TUB and PAG. A, Probability map of the proportion of participants that had a pathway between the TUB and PAG. B, 3D plot of the pathway shown in panel A. C, Probability map of the proportion of participants that had a pathway between PAG and whole brain gray matter. Olfactory areas including the anterior olfactory nucleus, frontal and temporal piriform cortex, TUB, medial amygdala, cortical amygdala, and the periamygdaloid complex were included as exclusion masks. TUB, olfactory tubercle; PAG, periaqueductal gray; L, left.

TUB–PAG tract integrity is correlated with BMI

We next sought to examine the behavioral significance of the fiber tract between the TUB and the PAG. We began by examining the PAG literature to look for potential PAG functions that could overlap with TUB functions. We found emerging evidence of a role for both the PAG and TUB in regulation of feeding behavior in rodents and humans (Jenck et al., 1987; Small et al., 2001; O’Doherty et al., 2006; Nolan-Poupart et al., 2013; Tryon and Mizumori, 2018; Nogi et al., 2020). Specifically, silencing local GABAergic cells in the ventrolateral PAG enhances feeding behavior, while activating them diminishes feeding behavior. There is a close relationship between olfaction and gustation, and a role for TUB in feeding behavior has been proposed (Murata, 2020). Combining these results, we hypothesized that the integrity of the tract between TUB and PAG would be inversely correlated with feeding behavior. The HCP dataset includes information about participant BMI, which can be used as an indirect measure of feeding behavior. We calculated the Pearson's coefficient between BMI and the fractional anisotropy of each voxel of the pathway between TUB and PAG. We found that BMI was negatively correlated with the average fractional anisotropy of the TUB–PAG pathway (Pearson's r = −0.28; p = 2.3 × 10−7) with no difference between female (Pearson's r = −0.31; p = 2.8 × 10−5) and male (Pearson's r = −0.24; p = 0.0028), as revealed by a direct comparison of r values (z = 0.67; p = 0.50; Fig. 7A). This correlation remained significant even when the FA of the whole-brain white matter skeleton was included as a covariate in the correlation analysis (Pearson's r = −0.26; p = 1.4 × 10−6; Fig. 8), suggesting specificity of the effects to the TUB–PAG pathway. Voxel-wise correlation analysis revealed that the TUB–PAG pathway has the strongest correlation with BMI in the anterior thalamic radiation (x = 5; y = −13; z = −4; Fig. 7B). This finding of a negative correlation between FA and BMI is in line with previous studies (Verstynen et al., 2012; Marqués-Iturria et al., 2015; Alarcón et al., 2016; Augustijn et al., 2018).

Figure 7.

Figure 7.

Open in a new tab

Correlation between BMI and fractional anisotropy of the TUB–PAG pathway. A, Scatterplot of the correlation between the average fractional anisotropy of the TUB–PAG pathway and BMI for females (black, n = 180) and males (gray, n = 155). The black and gray lines indicate fit lines. B, Left, spatial distribution of the correlation between fractional anisotropy of the TUB–PAG pathway and BMI (false discovery rate corrected p < 0.005). Right, Scatterplot of the maximal negative correlation (n = 335) at Montreal Neurological Institue coordinate (mm) x = 5; y = −13; z = −4. The black line indicates fit line at total. The pink outline indicates the group-level pathway. In panels A and B, each dot represents one participant. The correlation was examined using Pearson’s correlation coefficient. BMI, body mass index; TUB, olfactory tubercle; PAG, periaqueductal gray; R, right.

Figure 8.

Figure 8.

Open in a new tab

Correlation between BMI and FA of the pathway between periaqueductal gray and olfactory tubercle with the FA of whole brain white matter included as covariate. A, Scatterplot of the correlation between the average FA and BMI (n = 335). Each dot represents one participant, and the black line indicates fit line at total. B, Spatial distribution of the correlation between FA and BMI (false discovery rate corrected p < 0.005). The pink outline indicates the group-level pathway. The correlation was examined using Pearson’s correlation coefficient. BMI, body mass index; FA, fractional anisotropy; R, right.

Discussion

Food consumption is inherently rewarding when hungry, and thus reward circuitry plays an important role in regulation of food intake (Volkow et al., 2011). While the TUB receives monosynaptic input from the olfactory bulb, it is anatomically situated within the ventral striatum and exhibits multisensory properties (Wesson and Wilson, 2010), making it a potential region through which olfactory information could bias the motivation to eat (Wesson and Wilson, 2010; Murata, 2020). Here, we found a unique structural relationship between the TUB and PAG, such that TUB was the only olfactory bulbar cortical target and ventral striatal component that exhibited structural covariance with the PAG. We further found that among olfactory bulb targets, white matter streamlines with PAG were maximal in the TUB. Finally, we found a significant negative correlation between TUB–PAG tract integrity and BMI, which survived correction for whole-brain white matter FA. Overall, these findings suggest a unique relationship between the TUB and PAG which may be involved in feeding behavior. That said, our findings come with important limitations. Though our findings suggest specificity of effects to the TUB–PAG connection, constraints in current DTI techniques limit our ability to directly compare regions with overlapping white-matter tracts.

Given that BMI is an established proxy for eating in the absence of hunger (Lansigan et al., 2015; Savard et al., 2022), our findings suggest the need for future work focusing on a potential role for TUB in food intake regulation in humans, and specifically a possible role in obesity. Numerous feeding-related neuromodulators are highly and specifically expressed in the TUB, and odor-evoked TUB activity is modulated by appetite-regulating hormones (Zhao et al., 2023), broadly implicating TUB region in feeding behavior (Nogi et al., 2020). Histochemical studies conducted in rabbit pups (Olivo et al., 2014) and immunohistochemical studies conducted in bats (Chen et al., 2013) combine to suggest a potential role for the TUB in feeding behavior. However, the TUB remains relatively unexplored in humans, in terms of its structural organization, anatomical connectivity, and functionality.

The PAG is known as a key brain area in pain and threat processing; thus our findings linking this region with feeding behavior may be regarded as surprising. However, though most research on the PAG focuses on pain and threat, there is robust and growing evidence that the PAG may also be involved in consummatory behavior (Tryon and Mizumori, 2018), potentially through interactions between food reward and pain, due to its dual role in both (Geha et al., 2014). For example, administration of intraoral sucrose to rat pups elicits analgesia, an effect that is mediated by the PAG (Anseloni et al., 2005). Furthermore, a unique and robust positive association between responses to milkshakes in the PAG and ad libitum milkshake intake was identified in humans (Nolan-Poupart et al., 2013). Pharmacological inactivation of the PAG has been shown to reduce consummatory behavior in rodents and responses in PAG scale with reward size, suggesting a role for PAG in both feeding behavior and reward processing (Tryon and Mizumori, 2018). Optogenetic activation of local GABAergic cells in the ventrolateral PAG of rodents was found to suppress feeding behavior, while selective in vitro silencing of local GABAergic cells in the ventrolateral PAG enhanced feeding behavior (Hao et al., 2019), suggesting a role for the PAG in regulation of food intake. While the ventrolateral PAG is known to interact with the BNST and hypothalamus in regulation of feeding (Hao et al., 2019), we found that the integrity of anatomical connections between the PAG and the TUB is inversely associated with BMI. This suggests that connectivity between the TUB and PAG could potentially integrate olfactory information with feeding regulation circuits.

Multiple neural circuits are thought to be involved in feeding behavior (Small et al., 2001), which involves a range of complex processes such as motivation, search, discrimination, ingestion, and inhibition. Perception of food odors are not static in the human brain, but rather change with the smeller's state of hunger or satiety (Albrecht et al., 2009; Howard and Kahnt, 2017; Shanahan et al., 2021). It is therefore reasonable to expect that changes in food odor perception might be integrated into a neural circuit critical to changing feeding motivation in response to food need. Here, we show that the structural relationship between a the PAG and the TUB changes in step with BMI increases. BMI is a well-established surrogate measure of eating in the absence of hunger (Rothemund et al., 2007; Stoeckel et al., 2008; Stice et al., 2008a,b, 2010; Tomasi et al., 2009; Volkow et al., 2009; Batterink et al., 2010). Our findings could suggest that among the many circuits contributing to feeding behavior, a novel connection between the PAG and TUB is worth exploring in future work as a pathway that may convey an olfactory-specific satiety signal.

Our study has several limitations. Though BMI is significantly associated with food intake (Togo et al., 2001), it is not a direct measure and thus limits interpretation of our results. DTI methods have inherent limitations, including the inability to determine directionality of anatomical connections, and the inability to determine the number of synapses between target regions. Therefore, our findings cannot be interpreted to imply direct, monosynaptic connections between TUB and PAG, and we cannot infer whether PAG is modulating TUB responses or vice versa. We replicated findings in a second dataset where possible; however, one of the datasets did not include DTI or BMI, so we were only able to do this for the structural covariance findings.

In conclusion, we found a unique structural covariance between the TUB and PAG and that the integrity of the TUB–PAG pathway is negatively correlated with BMI. Our findings highlight a potential role for this pathway in the regulation of feeding behavior in humans.

References

  1. Alarcón G, Ray S, Nagel BJ (2016) Lower working memory performance in overweight and obese adolescents is mediated by white matter microstructure. J Int Neuropsychol Soc 22:281–292. 10.1017/S1355617715001265 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Albrecht J, Schreder T, Kleemann AM, Schöpf V, Kopietz R, Anzinger A, Demmel M, Linn J, Kettenmann B, Wiesmann M (2009) Olfactory detection thresholds and pleasantness of a food-related and a non-food odour in hunger and satiety. Rhinology 47:160–165. [PubMed] [Google Scholar]
  3. Allison AC (1954) The secondary olfactory areas in the human brain. J Anat 88:481–488. [PMC free article] [PubMed] [Google Scholar]
  4. Andersson J, Jenkinson M, Stephen S (2007) Non-linear registration aka spatial normalisation: FMRIB Technial Report TR07JA2. FMRIB Analysis Group of the University of Oxford. Available at: https://www.scienceopen.com/book?vid=13f3b9a9-6e99-4ae7-bea2-c1bf0af8ca6e [Accessed October 19, 2023].
  5. Anseloni VCZ, Ren K, Dubner R, Ennis M (2005) A brainstem substrate for analgesia elicited by intraoral sucrose. Neuroscience 133:231–243. 10.1016/j.neuroscience.2005.01.055 [DOI] [PubMed] [Google Scholar]
  6. Augustijn MJCM, Deconinck FJA, D’Hondt E, Van Acker L, De Guchtenaere A, Lenoir M, Caeyenberghs K (2018) Reduced motor competence in children with obesity is associated with structural differences in the cerebellar peduncles. Brain Imaging Behav 12:1000–1010. 10.1007/s11682-017-9760-5 [DOI] [PubMed] [Google Scholar]
  7. Batterink L, Yokum S, Stice E (2010) Body mass correlates inversely with inhibitory control in response to food among adolescent girls: an fMRI study. Neuroimage 52:1696–1703. 10.1016/j.neuroimage.2010.05.059 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Behrens TE, et al. (2003) Non-invasive mapping of connections between human thalamus and cortex using diffusion imaging. Nat Neurosci 6:750–757. 10.1038/nn1075 [DOI] [PubMed] [Google Scholar]
  9. Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Series B Stat Methodol 57:289–300. 10.1111/j.2517-6161.1995.tb02031.x [DOI] [Google Scholar]
  10. Bhutani S, Howard JD, Reynolds R, Zee PC, Gottfried J, Kahnt T (2019) Olfactory connectivity mediates sleep-dependent food choices in humans. Elife 8:e49053. 10.7554/eLife.49053 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Buckner RL, Head D, Parker J, Fotenos AF, Marcus D, Morris JC, Snyder AZ (2004) A unified approach for morphometric and functional data analysis in young, old, and demented adults using automated atlas-based head size normalization: reliability and validation against manual measurement of total intracranial volume. Neuroimage 23:724–738. 10.1016/j.neuroimage.2004.06.018 [DOI] [PubMed] [Google Scholar]
  12. Cansler HL, Wright KN, Stetzik LA, Wesson DW (2020) Neurochemical organization of the ventral striatum’s olfactory tubercle. J Neurochem 152:425–448. 10.1111/jnc.14919 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Chen Q, Wang L, Jones G, Metzner W, Xuan FJ, Yin J, Sun Y (2013) Foxp2 and olfaction: divergence of FoxP2 expression in olfactory tubercle between different feeding habit bats. Acta Biol Hung 64:426–437. 10.1556/ABiol.64.2013.4.3 [DOI] [PubMed] [Google Scholar]
  14. Coifman RR, Lafon S, Lee AB, Maggioni M, Nadler B, Warner F, Zucker SW (2005) Geometric diffusions as a tool for harmonic analysis and structure definition of data: diffusion maps. Proc Natl Acad Sci U S A 102:7426–7431. 10.1073/pnas.0500334102 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Douaud G, et al. (2007) Anatomically related grey and white matter abnormalities in adolescent-onset schizophrenia. Brain 130:2375–2386. 10.1093/brain/awm184 [DOI] [PubMed] [Google Scholar]
  16. Gadziola MA, Tylicki KA, Christian DL, Wesson DW (2015) The olfactory tubercle encodes odor valence in behaving mice. J Neurosci 35:4515–4527. 10.1523/JNEUROSCI.4750-14.2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Geha P, deAraujo I, Green B, Small DM (2014) Decreased food pleasure and disrupted satiety signals in chronic low back pain. Pain 155:712–722. 10.1016/j.pain.2013.12.027 [DOI] [PubMed] [Google Scholar]
  18. George DT, Ameli R, Koob GF (2019) Periaqueductal gray sheds light on dark areas of psychopathology. Trends Neurosci 42:349–360. 10.1016/j.tins.2019.03.004 [DOI] [PubMed] [Google Scholar]
  19. Glasser MF, et al. (2013) The minimal preprocessing pipelines for the Human Connectome Project. Neuroimage 80:105–124. 10.1016/j.neuroimage.2013.04.127 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Good CD, Johnsrude IS, Ashburner J, Henson RN, Friston KJ, Frackowiak RS (2001) A voxel-based morphometric study of ageing in 465 normal adult human brains. Neuroimage 14:21–36. 10.1006/nimg.2001.0786 [DOI] [PubMed] [Google Scholar]
  21. Hao S, et al. (2019) The lateral hypothalamic and BNST GABAergic projections to the anterior ventrolateral periaqueductal gray regulate feeding. Cell Rep 28:616–624.e5. 10.1016/j.celrep.2019.06.051 [DOI] [PubMed] [Google Scholar]
  22. Howard JD, Gottfried JA (2014) Configural and elemental coding of natural odor mixture components in the human brain. Neuron 84:857–869. 10.1016/j.neuron.2014.10.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Howard JD, Kahnt T (2017) Identity-specific reward fepresentations in orbitofrontal cortex are modulated by selective devaluation. J Neurosci 37:2627–2638. 10.1523/JNEUROSCI.3473-16.2017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Ikemoto S (2003) Involvement of the olfactory tubercle in cocaine reward: intracranial self-administration studies. J Neurosci 23:9305–9311. 10.1523/JNEUROSCI.23-28-09305.2003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Ikemoto S (2007) Dopamine reward circuitry: two projection systems from the ventral midbrain to the nucleus accumbens-olfactory tubercle complex. Brain Res Rev 56:27–78. 10.1016/j.brainresrev.2007.05.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Jenck F, Quirion R, Wise RA (1987) Opioid receptor subtypes associated with ventral tegmental facilitation and periaqueductal gray inhibition of feeding. Brain Res 423:39–44. 10.1016/0006-8993(87)90822-5 [DOI] [PubMed] [Google Scholar]
  27. Jenkinson M, Beckmann CF, Behrens TEJ, Woolrich MW, Smith SM (2012) FSL. Neuroimage 62:782–790. 10.1016/j.neuroimage.2011.09.015 [DOI] [PubMed] [Google Scholar]
  28. Johansen-Berg H, Behrens TEJ, Robson MD, Drobnjak I, Rushworth MFS, Brady JM, Smith SM, Higham DJ, Matthews PM (2004) Changes in connectivity profiles define functionally distinct regions in human medial frontal cortex. Proc Natl Acad Sci U S A 101:13335–13340. 10.1073/pnas.0403743101 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Keuken MC, Forstmann BU (2015) A probabilistic atlas of the basal ganglia using 7T MRI. Data Brief 4:577–582. 10.1016/j.dib.2015.07.028 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kharabian Masouleh S, Plachti A, Hoffstaedter F, Eickhoff S, Genon S (2020) Characterizing the gradients of structural covariance in the human hippocampus. Neuroimage 218:116972. 10.1016/j.neuroimage.2020.116972 [DOI] [PubMed] [Google Scholar]
  31. Lansigan RK, Emond JA, Gilbert-Diamond D (2015) Understanding eating in the absence of hunger among young children: a systematic review of existing studies. Appetite 85:36–47. 10.1016/j.appet.2014.10.032 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Leclercq S, Blancher G (2012) Multimodal sensory integration during sequential eating–linking chewing activity, aroma release, and aroma perception over time. Chem Senses 37:689–700. 10.1093/chemse/bjs047 [DOI] [PubMed] [Google Scholar]
  33. Linnman C, Moulton EA, Barmettler G, Becerra L, Borsook D (2012) Neuroimaging of the periaqueductal gray: state of the field. Neuroimage 60:505–522. 10.1016/j.neuroimage.2011.11.095 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Mai JK, Majtanik M, Paxinos G (2015) Atlas of the human brain, Ed 4. Academic Press. [Google Scholar]
  35. Marqués-Iturria I, et al. (2015) Affected connectivity organization of the reward system structure in obesity. Neuroimage 111:100–106. 10.1016/j.neuroimage.2015.02.012 [DOI] [PubMed] [Google Scholar]
  36. Midroit M, et al. (2021) Neural processing of the reward value of pleasant odorants. Curr Biol 31:1592–1605.e9. 10.1016/j.cub.2021.01.066 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Millman DJ, Murthy VN (2020) Rapid learning of odor-value association in the olfactory striatum. J Neurosci 40:4335–4347. 10.1523/JNEUROSCI.2604-19.2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Murata K (2020) Hypothetical roles of the olfactory tubercle in odor-guided eating behavior. Front Neural Circuits 14:577880. 10.3389/fncir.2020.577880 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Murata K, Kanno M, Ieki N, Mori K, Yamaguchi M (2015) Mapping of learned odor-induced motivated behaviors in the mouse olfactory tubercle. J Neurosci 35:10581–10599. 10.1523/JNEUROSCI.0073-15.2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Nogi Y, Ahasan MM, Murata Y, Taniguchi M, Sha MFR, Ijichi C, Yamaguchi M (2020) Expression of feeding-related neuromodulatory signalling molecules in the mouse central olfactory system. Sci Rep 10:890. 10.1038/s41598-020-57605-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Nolan-Poupart S, Veldhuizen MG, Geha P, Small DM (2013) Midbrain response to milkshake correlates with ad libitum milkshake intake in the absence of hunger. Appetite 60:168–174. 10.1016/j.appet.2012.09.032 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Noto T, Zhou G, Yang Q, Lane G, Zelano C (2021) Human primary olfactory amygdala subregions form distinct functional networks, suggesting distinct olfactory functions. Front Syst Neurosci 15:752320. 10.3389/fnsys.2021.752320 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. O’Doherty JP, Buchanan TW, Seymour B, Dolan RJ (2006) Predictive neural coding of reward preference involves dissociable responses in human ventral midbrain and ventral striatum. Neuron 49:157–166. 10.1016/j.neuron.2005.11.014 [DOI] [PubMed] [Google Scholar]
  44. O’Doherty J, Rolls ET, Francis S, Bowtell R, McGlone F, Kobal G, Renner B, Ahne G (2000) Sensory-specific satiety-related olfactory activation of the human orbitofrontal cortex. Neuroreport 11:893–897. 10.1097/00001756-200003200-00046 [DOI] [PubMed] [Google Scholar]
  45. Olivo D, Caba M, Gonzalez-Lima F, Vázquez A, Corona-Morales A (2014) Circadian feeding entrains anticipatory metabolic activity in piriform cortex and olfactory tubercle, but not in suprachiasmatic nucleus. Brain Res 1592:11–21. 10.1016/j.brainres.2014.09.054 [DOI] [PubMed] [Google Scholar]
  46. Omelchenko N, Sesack SR (2010) Periaqueductal gray afferents synapse onto dopamine and GABA neurons in the rat ventral tegmental area. J Neurosci Res 88:981–991. 10.1002/jnr.22265 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Rolls ET (2015) Taste, olfactory, and food reward value processing in the brain. Prog Neurobiol 127–128:64–90. 10.1016/j.pneurobio.2015.03.002 [DOI] [PubMed] [Google Scholar]
  48. Rothemund Y, Preuschhof C, Bohner G, Bauknecht H-C, Klingebiel R, Flor H, Klapp BF (2007) Differential activation of the dorsal striatum by high-calorie visual food stimuli in obese individuals. Neuroimage 37:410–421. 10.1016/j.neuroimage.2007.05.008 [DOI] [PubMed] [Google Scholar]
  49. Savard C, Bégin S, Gingras V (2022) Factors associated with eating in the absence of hunger among children and adolescents: a systematic review. Nutrients 14:4715. 10.3390/nu14224715 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Schwob JE, Price JL (1984) The development of axonal connections in the central olfactory system of rats. J Comp Neurol 223:177–202. 10.1002/cne.902230204 [DOI] [PubMed] [Google Scholar]
  51. Shanahan LK, Bhutani S, Kahnt T (2021) Olfactory perceptual decision-making is biased by motivational state. PLoS Biol 19:e3001374. 10.1371/journal.pbio.3001374 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Shin S, You I-J, Jeong M, Bae Y, Wang X-Y, Cawley ML, Han A, Lim BK (2023) Early adversity promotes binge-like eating habits by remodeling a leptin-responsive lateral hypothalamus-brainstem pathway. Nat Neurosci 26:79–91. 10.1038/s41593-022-01208-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Small DM, Zatorre RJ, Dagher A, Evans AC, Jones-Gotman M (2001) Changes in brain activity related to eating chocolate: from pleasure to aversion. Brain 124:1720–1733. 10.1093/brain/124.9.1720 [DOI] [PubMed] [Google Scholar]
  54. Smith SM, Nichols TE (2009) Threshold-free cluster enhancement: addressing problems of smoothing, threshold dependence and localisation in cluster inference. Neuroimage 44:83–98. 10.1016/j.neuroimage.2008.03.061 [DOI] [PubMed] [Google Scholar]
  55. Stice E, Spoor S, Bohon C, Small DM (2008a) Relation between obesity and blunted striatal response to food is moderated by TaqIA A1 allele. Science 322:449–452. 10.1126/science.1161550 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Stice E, Spoor S, Bohon C, Veldhuizen MG, Small DM (2008b) Relation of reward from food intake and anticipated food intake to obesity: a functional magnetic resonance imaging study. J Abnorm Psychol 117:924–935. 10.1037/a0013600 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Stice E, Yokum S, Bohon C, Marti N, Smolen A (2010) Reward circuitry responsivity to food predicts future increases in body mass: moderating effects of DRD2 and DRD4. Neuroimage 50:1618–1625. 10.1016/j.neuroimage.2010.01.081 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Stoeckel LE, Weller RE, Cook EW, Twieg DB, Knowlton RC, Cox JE (2008) Widespread reward-system activation in obese women in response to pictures of high-calorie foods. Neuroimage 41:636–647. 10.1016/j.neuroimage.2008.02.031 [DOI] [PubMed] [Google Scholar]
  59. Togo P, Osler M, Sørensen TI, Heitmann BL (2001) Food intake patterns and body mass index in observational studies. Int J Obes Relat Metab Disord 25:1741–1751. 10.1038/sj.ijo.0801819 [DOI] [PubMed] [Google Scholar]
  60. Tomasi D, Wang G-J, Wang R, Backus W, Geliebter A, Telang F, Jayne MC, Wong C, Fowler JS, Volkow ND (2009) Association of body mass and brain activation during gastric distention: implications for obesity. PLoS One 4:e6847. 10.1371/journal.pone.0006847 [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Tripathi A, Prensa L, Mengual E (2013) Axonal branching patterns of ventral pallidal neurons in the rat. Brain Struct Funct 218:1133–1157. 10.1007/s00429-012-0451-0 [DOI] [PubMed] [Google Scholar]
  62. Tryon VL, Mizumori SJY (2018) A novel role for the periaqueductal gray in consummatory behavior. Front Behav Neurosci 12:178. 10.3389/fnbeh.2018.00178 [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Van Essen DC, Smith SM, Barch DM, Behrens TEJ, Yacoub E, Ugurbil K, WU-Minn HCP Consortium (2013) The WU-Minn Human Connectome Project: an overview. Neuroimage 80:62–79. 10.1016/j.neuroimage.2013.05.041 [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Verhagen JV, Engelen L (2006) The neurocognitive bases of human multimodal food perception: sensory integration. Neurosci Biobehav Rev 30:613–650. 10.1016/j.neubiorev.2005.11.003 [DOI] [PubMed] [Google Scholar]
  65. Verstynen TD, Weinstein AM, Schneider WW, Jakicic JM, Rofey DL, Erickson KI (2012) Increased body mass index is associated with a global and distributed decrease in white matter microstructural integrity. Psychosom Med 74:682–690. 10.1097/PSY.0b013e318261909c [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Vertes RP, Linley SB, Hoover WB (2015) Limbic circuitry of the midline thalamus. Neurosci Biobehav Rev 54:89–107. 10.1016/j.neubiorev.2015.01.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Volkow ND, et al. (2009) Inverse association between BMI and prefrontal metabolic activity in healthy adults. Obesity 17:60–65. 10.1038/oby.2008.469 [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Volkow ND, Wang G-J, Baler RD (2011) Reward, dopamine and the control of food intake: implications for obesity. Trends Cogn Sci 15:37–46. 10.1016/j.tics.2010.11.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Vos de Wael R, et al. (2020) Brainspace: a toolbox for the analysis of macroscale gradients in neuroimaging and connectomics datasets. Commun Biol 3:103. 10.1038/s42003-020-0794-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Wesson DW (2020) The tubular striatum. J Neurosci 40:7379–7386. 10.1523/JNEUROSCI.1109-20.2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Wesson DW, Wilson DA (2010) Smelling sounds: olfactory-auditory sensory convergence in the olfactory tubercle. J Neurosci 30:3013–3021. 10.1523/JNEUROSCI.6003-09.2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Zhang Z, Zhang H, Wen P, Zhu X, Wang L, Liu Q, Wang J, He X, Wang H, Xu F (2017) Whole-brain mapping of the inputs and outputs of the medial part of the olfactory tubercle. Front Neural Circuits 11:52. 10.3389/fncir.2017.00052 [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Zhao Y, Bhutani S, Kahnt T (2023) Appetite-regulating hormones modulate odor perception and odor-evoked activity in hypothalamus and olfactory cortices. Chem Senses 48:bjad039. 10.1093/chemse/bjad039 [DOI] [PMC free article] [PubMed] [Google Scholar]
  74. Zhou G, Lane G, Cooper SL, Kahnt T, Zelano C (2019) Characterizing functional pathways of the human olfactory system. Elife 8:e47177. 10.7554/eLife.47177 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The source data related to this study are available at https://github.com/zelanolab/human_olfactory_tubercle_and_periaqueductal_gray_structural_connectivity.

Articles from The Journal of Neuroscience are provided here courtesy of Society for Neuroscience

RESOURCES