Imaging the three human primary amygdala output tracts with age and sex characterisation across the lifespan

doi:10.21203/rs.3.rs-4612085/v1

Download PDF

Research Article

Imaging the three human primary amygdala output tracts with age and sex characterisation across the lifespan

https://doi.org/10.21203/rs.3.rs-4612085/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

- The amygdala is involved memory and emotion processing and communicates with the rest of the brain through three efferent tracts: the stria terminalis (ST), ventral amygdalofugal pathway (VAP), and anterior commissure (AC). This is the first study to isolate all three major amygdalar output pathways using novel diffusion tractography protocols with exploration of the diffusion, age, and sex characteristics of each tract.

Methods

− 64 healthy individuals aged 15–64 underwent high-resolution T1, T2, and diffusion MR brain imaging with constrained spherical deconvolution tractography. Individually generated amygdalae aided the virtual dissection of the ST, VAP and AC using novel bespoke protocols for each tract based on anatomical principles. Age and sex diffusion characteristics were explored.

Results

- The ST showed age-related decreases in fractional anisotropy (left: p = 0.00018; right: p = 0.00032), mean diffusivity (left: p = 0.0017; right: p = 0.00058), and radial diffusivity (left: p = 0.00015; right: p = 3.44E-05). The AC showed decreases in mean diffusivity (p = 0.0022) and axial diffusivity (p = 0.00015). Sex had no significant effect on the diffusion metrics apart from the right ST, showing higher fractional anisotropy in males than in females (p = 0.001).

Conclusion

- This is the first study to virtually dissect the three main output tracts of the amygdala from neuroimaging. We also show age related changes in markers of neuronal integrity with age. No sex differences were found apart from potentially more robust integrity in the right ST in males. The novel anatomically-driven and reproducible protocols for ST and VAP isolation presented may guide future investigation of the connectivity and efferent circuitry of the amygdala.

Amygdala

Stria Terminalis

Amygdalofugal Pathway

Neuroimaging

Anterior Commissure

Embedded within the medial temporal lobe of the brain is the amygdala; an almond-shaped assembly of nuclei that is known to be involved in various limbic processes such as fear, reward, emotional learning and aggression (AbuHasan, Reddy, & Siddiqui, 2020; Amaral, 2003; Amunts et al., 2005; Michael Davis, 1994; Duvarci & Pare, 2014; Haller, 2018; Stamatakis et al., 2014). The amygdala has been shown to be involved in various neuropsychiatric illnesses from depression (Nolan et al., 2020; D. Roddy et al., 2021) to psychosis (O'Neill et al., 2022; O’Neill et al., 2024). Given the fundamental nature of the amygdala to a diversity of brain functions and it’s interactions with the majority of cortical and subcortical regions, it is important to consider the efferent connections through which the amygdala communicates information to the rest of the brain.

Classically, the three primary white matter efferent tracts from the amygdala are the stria terminalis (ST), ventral amygdalofugal pathway (VAP) and anterior commissure (AC) (Hsu, Huang, & Hsueh, 2020; Noback, Ruggiero, Demarest, & Strominger, 2005; Stamatakis et al., 2014) (Fig. 1). When combined with the bed nucleus of the stria terminalis (BNST), the central nucleus of the amygdala forms core components of the central extended amygdala (M. Davis & Shi, 1999). This central extended amygdala is continuous with the ST dorsally and with the VAP ventrally (Alheid, 2003; Cassell, Freedman, & Shi, 1999). The anatomy of each of the three main output tracts of the amygdala is closely related to their function (Gray, 1999). While the ST travels in a C-shaped path closely against the ventricular surface of the thalamus alongside the arc of the fornix, the VAP directly connects the amygdala to the hypothalamus ventrally. Traversing between the anterior and posterior columns of the fornix, the AC connects the left and right temporal lobes to allow for interhemispheric communication (Winter & Franz, 2014).

The ST consists of a narrow band of white matter that run along the ventricular surface of the thalamus (Koller, Hatton, Rogers, & Rafal, 2019). It allows for communication between the amygdala and hypothalamus, amygdala and basal forebrain. For the majority of its course, the axons are tightly aligned with the fornix and follow a C-shaped path in humans (Lee & Davis, 1997). Similar to the fornix, ST fibres split across the AC to those fibres going anteriorly to the basal forebrain (precommissural fibres) and those fibres going posteriorly to the hypothalamus (postcommisural fibres). The close proximity of the ST to the fornix has historically made it challenging for neuroimaging researchers to examine the ST as an independent entity (Baydin et al., 2017; Pascalau, Popa Stanila, Sfrangeu, & Szabo, 2018). The VAP emanates from the amygdala ventrally towards the ST and connects to the hypothalamus via a relatively direct route (Kamali et al., 2016). The VAP gives off branches which connect with the basal forebrain, mediodorsal thalamus, nucleus accumbens and the brainstem (Miller, Saint Marie, Breier, & Swerdlow, 2010; Porrino, Crane, & Goldman-Rakic, 1981).

While the ST and VAP project to similar areas of the brain (the hypothalamus and basal forebrain), regions important for stress responses and pleasure/motivation respectively, the ST takes a more indirect route to reach them (Bao & Swaab, 2019; Berridge & Kringelbach, 2015; Kamali et al., 2016). When observed histologically using luxol fast blue stained cross-sections, the VAP is also seen to contain a greater proportion of myelinated axons, appearing more hyper-chromatic than the ST (Mori et al., 2017). Greater myelination means greater insulation and therefore, faster signal transmission (Nave & Werner, 2014). Therefore, the VAP connections to the hypothalamus/basal forebrain are considered more direct and ‘faster’, whereas the ST connections to the hypothalamus/basal forebrain take a longer, more convoluted course. This is mirrored in the nomenclature, as the term fugal is derived from the Latin fugiō: “to flee”. Both structures allow the amygdala to modulate homeostatic/stress responses (hypothalamic connections) as well as attention and arousal (basal forebrain). The AC links both temporal lobes together. It traverses the midline between the anterior and posterior columns of the fornix and ST (Kiernan, 2012; Mai, Majtanik, & Paxinos, 2015; Winter & Franz, 2014), splitting both tracts into precommissural (basal forebrain) and postcommisural (hypothalamic) fibres. The AC allows the amygdalae to communicate across the midline.

Diffusion-weighted imaging (DWI) is sensitive to the motion of water molecules in tissues, which is restricted by tissue boundaries, and therefore, allows for a quantitative means to describe tissue microstructural characteristics (Le Bihan et al., 2001). From DWI, white matter tracts can be reconstructed and visualised in-vivo and non-invasively (Basser, Pajevic, Pierpaoli, Duda, & Aldroubi, 2000). Recent developments in DWI permit even greater precision of tract isolation (Johansen-Berg & Rushworth, 2009) (Farquharson et al., 2013). To allow for better tract delineation, these advancements first look to improve the signal-to-noise ratio (SNR). This is achieved through use of higher MR field strengths and faster acquisition sequences to reduce scan time (J. M. Soares, P. Marques, V. Alves, & N. Sousa, 2013). Improvements in pre-processing techniques, such as correcting for head motion, distortion, and free water, also contribute to an improved SNR (Tournier, Mori, & Leemans, 2011). Additionally, models like constrained spherical deconvolution (Jeurissen, Leemans, Jones, Tournier, & Sijbers, 2011), enable the exploration of previously obscured crossing, diverging, and kissing fibres. As aforementioned, the close proximity of the fornix and the ST has previously made ST delineation difficult, however, the above DWI advances have allowed for more accurate imaging and therefore, more precise tract reconstruction, particularly of complex limbic tracts (Kamali et al., 2015; Nasa et al., 2021; Darren William Roddy et al., 2022).

Prior research into the three amygdalar efferents using DWI tractography is scant. This study aims to use an anatomically driven approach to delineate these three outflow tracts from diffusion MRI images for the first time in a normal population. In particular, a delineation of the major branches of the VAP and a novel method for separating out the ST from its adjacent fornix will be presented for the first time. This study also explores the relationship of sex and age on the microstructural diffusion properties of these white matter tracts.

Participants

Sixty-four subjects were enrolled through an active database of willing participants as part of the REDEEM (Research in Depression: Endocrinology, Epigenetics, and neuroiMaging) research study at Trinity College Dublin. Exclusion criteria for study entry included contraindications to MRI, history of illicit substance abuse, head trauma, and any significant medical or psychiatric illness. All participants underwent a Structured Clinical Interview for DSM-IV (SCID) and were subsequently given scores on the Hamilton Depression Rating Scale (HAM-D) and the Hamilton Anxiety Rating Scale (HAM-A) (Max Hamilton, 1959; M. Hamilton, 1960). The interview and scoring were performed by a psychiatrist at Trinity College Institute of Neuroscience. Inclusion in the study required subjects to have no active or previous SCID diagnosis of depression or anxiety, along with a HAM-D score of less than 9 and a HAM-A score of less than 14.

This study was carried out in accordance with the recommendations of the Tallaght Hospital/St. James Hospital Joint Research Ethics Committee. All subjects gave written informed consent in accordance with the Declaration of Helsinki. The protocol was approved by the Tallaght Hospital/St. James Hospital Joint Research Ethics Committee.

MRI Data Acquisition

All data were acquired on a Philips Intera Achieva 3.0T magnetic resonance system (32-channel head coil) at Trinity College Institute of Neuroscience, Dublin. As a part of this study, three imaging sequences were acquired: T1, T2-FLAIR and whole brain High Angular Resolution Diffusion Imaging (HARDI) (Jones, Knösche, & Turner, 2013; Jones & Leemans, 2011). The scan parameters were as follows:

T1: 180 axial high-resolution T1-weighted anatomical images (T1W-IR1150 sequence, echo time = 3.8 ms, repetition time = 8.4 ms, field of view 230 mm, 0.898 x 0.898 mm² in-plane resolution, slice thickness 0.9 mm, flip angle α = 8°).

T2-FLAIR: 60 axial T2-FLAIR-weighted images (echo time = 120 ms, repetition time = 2800 ms, field of view = 230 mm, 0.49 x 0.49 mm² in-plane resolution, slice thickness = 3 mm, flip angle α = 8°). Slices were taken in the axial plane, with the resulting smaller in-plane resolution corresponding to the longitudinal axis of the amygdala.

Whole brain HARDI: 60 axial HARDI images were acquired in 61 non-collinear gradient directions using a spin-echo echo-planar imaging (SE EPI) pulse sequence (echo time = 52 ms, repetition time = 11,260 ms, field of view = 224 mm, 2mm³ isotropic voxels, b-value = 1500smm^− 2, flip angle α = 90°).

Image Analyses Using FreeSurfer 6.0

Cortical reconstruction and segmentation were performed using the FreeSurfer 6.0 image analysis suite (Fischl, 2012). FreeSurfer interrogates contrast differences using previously defined in-vivo and ex-vivo amygdala atlases to determine the 3D boundaries and volume of the amygdala. The procedure was optimised by combining T1 and T2-FLAIR-weighted inputs and selecting the 3T MRI flag and multispectral segmentation in FreeSurfer. Similarly, Freesurfer 6.0 also generated a nucleus accumbens for VAP branch isolation (see below). An estimate of total intracranial volume was also calculated for each subject.

DWI Pre-processing and Whole Brain Tractography

Pre-processing of Diffusion-Weighted Imaging (DWI) was done using ExploreDTI, a free MATLAB-2 based graphical toolbox for diffusion MRI pre-processing and tractography (Leemans, Jeurissen, Sijbers, & Jones, 2009). This pre-processing involved the following steps:

(a) Exporting and standardizing diffusion output files

(b) Signal drift correction via linear and quadratic correction (Vos et al., 2017)

(d) Orientation/Directionality checks via manual glyph inspection

(e) Head motion and eddy current artefact correction using rigid body and affine registration respectively to the non-diffusion weighted b₀ image (J. Soares, P. Marques, V. Alves, & N. Sousa, 2013)

(f) EPI deformation correction via affine registration to the T1 image (Irfanoglu, Sarlls, Nayak, & Pierpaoli, 2019)

Each of the above pre-processing steps were performed using the latest version toolbox plugins directly within ExploreDTI.

Tractography of the whole brain was generated in ExploreDTI using constrained spherical deconvolution as a recursive calibration of the response function (Jeurissen et al., 2011). Whole brain seeding was used with seed voxel sizes of 2 mm³ and a fibre orientation distribution threshold of 0.1. Whole brain tractography was achieved through multiple random seed placements of one seed per voxel (Jones & Leemans, 2011). The step size was set as 0.5mm and the maximum angle threshold was set as 89°. All whole brain tract lengths were set between 10 and 500 mm. Previous diffusion studies further detail the above steps for DWI-pre-processing and whole brain tractography (Gaughan et al., 2023; Nasa et al., 2021; Darren W Roddy et al., 2018).

Tract isolation

Following consultation with the Department of Anatomy at Trinity College Dublin, Boolean logic protocols were designed to isolate the amygdalar outputs within ExploreDTI. Emphasis was given towards Region of Interests (ROI) that included the maximum possible number of streamlines; were positioned around landmarks clearly identifiable on HARDI or T1-weighted images and that would consistently capture the desired tract. Following an initial training period on a different dataset of 10 subjects, two raters placed the gates on DW and T1 MR images. Raters were blind to age and sex of the subjects. A comparison of reliability between each independent rater was performed using interclass correlation coefficient (ICC) analysis using SPSS26 for the following measurements: dimensional measures, length volume and standard diffusion metrics; fractional anisotropy, mean diffusivity, axial diffusivity and radial diffusivity. Tract were reviewed visually by an independent rater and the isolated repeated/adjusted if the tract did not achieve an ICC of > 0.9 between raters. Disputes between raters were resolved by a consensus meeting between the two raters and two independent raters with experience in neuroanatomy and neuroimaging.

To isolate the ST

Initially, the fornix/stria terminalis were isolated together as a unit. This was done since the two tracts are tightly aligned and the manual placement of ROIs to segregate them could lead to either inappropriately excluding fibres from the ST or inappropriately including fibres from the fornix within the ST tract. This initial approach involved the placement of six gates. Two of these were ROI AND gates and the other four were Regions of Avoidance (ROA). Then, the amygdalar masks) generated using Freesurfer 6.0 were used as three-dimensional AND gates for each side separately to segregate the fornical streamlines from the combination tract. This resulted in the isolation of the left and right ST. Tracts were confirmed visually to correspond to the typical arc-like shape of the ST, and cleaned for extraneous streamlines. The complete ST from the amygdala to the columns of the ST as they straddle the AC was used for analysis.

To isolate the AC

A ROI was placed as an AND gate around the anterior commissure on the sagittal plane in the midline. Tracts were confirmed visually to intersect with both amygdalae. The AC was cleaned for extraneous streamlines and a core section between the two amygdale used for analysis.

To isolate the VAP

The amygdalar masks generated using Freesurfer 6.0 were used as three-dimensional AND gates to remove all of the streamlines that didn’t reach the amygdala. This was an important step for increasing the signal-to-noise ratio in the region. Following this, manual placement of gates was conducted on the left and right side separately. For the left side, an AND gate was placed in the sagittal plane directly medial to the ipsilateral amygdala. The gate was placed around the landmark for AC, optic tract and the VAP. This captured much noise but was a measure to ensure that the subtle branches of the VAP would not mistakenly get removed. Next, a NOT gate was placed on the entire coronal slice directly posterior to the ipsilateral amygdala. This reduced noise in the area and allowed the raters placing the gates to visualize the region more clearly. Then, concentric OR gates were placed around potential hypothalamic, basal forebrain, mediodorsal thalamic and nucleus accumbens branches of the tract using directionality towards these regions as a guide. The resulting tract from all of the above steps was the VAP.

The three of the branches of the VAP were also isolated in this study. To isolate the fibres emanating towards the hypothalamus, an AND gate was placed medial to the VAP where the hypothalamic branch could be visualized. A Freesurfer-generated volume mask of the nucleus accumbens was used as an AND gate to isolate the VAP branch extending towards the nucleus accumbens (Fig. 2). Finally, the branch extending to the basal forebrain was isolated by placing an AND gate infero-medial to the VAP. The fibres going towards the mediodorsal thalamus were unable to be consistently isolated. This branch appeared to involve sparse and inconsistent streamlines.

Cleaning and Segmentation

NOT gates were used to remove extraneous streamlines which arose in resultant tracts either due to over-accommodation by the gates from the study protocol or by miscalculation due to proximity of these extraneous fibres to these gates.

Each of the resulting tracts were segmented systematically. This was so that statistical analysis could later be implemented on set regions. The AC was segmented to keep six MRI slices of the central portion of the tract. The main body of the ST was segmented. For the VAP, four MRI slices of the tract were segmented. The three VAP branches were also segmented separately.

Statistical Analysis

All analyses were performed using SPSS-26 ("IBM Corp. Released 2019. IBM SPSS Statistics for Windows, Version 26.0. Armonk, NY: IBM Corp,").

One-way between-group analyses of covariance (ANCOVA) were used to compare mean differences in the diffusion metrics of each of the amygdalar output tracts between males and females. Age and estimated total intracranial volume (eTIV) were inputted as covariates.. This was performed separately for each diffusion metric for all 3 output tracts. Prior to conducting this test, the assumption tests for an ANCOVA were conducted to check for normality, linearity, univariate outliers, homogeneity of variances, and homogeneity of regression slopes between the covariates and the dependent variables.

In addition, partial correlations were also performed in order to identify and quantify any potential correlation between age and the investigated diffusion metrics while controlling for the effects of gender and eTIV (Table 1).

Significance was determined to be a p-value of less than 0.05 and the effect size was measured by partial eta² ( < = 0.01: low effect size, 0.09–0.25: moderate effect size, >=0.25: large effect size). Bonferroni correction, a conservative correction of type 1 errors, was applied to these analyses. This further adjusted the level to reach statistical significance from 0.05 to 0.0033 on the basis of 15 separate analyses (3 eigenvectors analysed across the tracts in both cerebral hemispheres).

Table 1

Mean age and estimated total intracranial volume (eTIV) compared between males and females. eTIV = estimated total intracranial volume.
	Males (n = 23)	Females (n = 41)	Total (n = 64)
	Mean ± SD	Mean ± SD	Mean ± SD
Age	30.3 ± 14.5	31.7 ± 12.3	31.2 ± 13
eTIV (mm3)	1652765 ± 128139	1472352 ± 1472352	1537188 ± 143084

The ST, AC and VAP were generated consistently for all subjects. Three branches of the VAP were identified and isolated: a branch to the basal forebrain, a branch to the hypothalamus and a branch to the nucleus accumbens. Tract data from the metrically constrained segments was statistically analysed to seek differences and changes in these tracts with age and between sex.

Following correction for sex and eTIV, partial correlations showed that the coefficient of determination (r²), which measures the linear relationship of two variables by highlighting how differences in one variable can be explained by the differences in another variable, was significant in both the left and right ST as well as in the AC. As shown in Table 2, the left and right ST showed a decrease in fractional anisotropy (left p = 0.00018, r² = 0.21; right p = 0.00032, r² = 0.19), a decrease in mean diffusivity (left p = 0.0017, r² = 0.15; right p = 0.00058, r² = 0.18) and a decrease in radial diffusivity (left p = 0.00015, r² = 0.21; right p = 3.44E-05, r² = 0.25) as the age of the subjects increased. The AC also showed a decrease in mean diffusivity (p = 0.0022, r² = 0.15) as well as a decrease in axial diffusivity (p = 0.00015, r² = 0.21) with increasing age, with no correlation being observed for either fractional anisotropy or radial diffusivity. It is important to note that the anterior commissure is not a bilateral structure, i.e. the human brain contains a single AC, which explains why the AC only has one set of results. No correlations were observed for the left or right VAP.

Table 2

Correlations between various diffusion metrics and age for the left and right stria terminalis, the left and right ventral amygdalofugal pathway and the anterior commissure. Corrections made for gender and estimated total intracranial volume. Statistically significant values (p < 0.05) are highlighted in bold. r = correlation coefficient
		Stria Terminalis		Ventral Amygdalofugal Pathway		Anterior Commissure
		Left	Right	Left	Right	Anterior Commissure
Fractional Anisotropy	r	-0.46144	-0.44186	-0.06500	0.28900	0.09235
	p value	0.00018	0.00032	0.61600	0.02300	0.47531
Mean Diffusivity	r	0.39240	0.42797	0.05800	-0.15900	0.38189
	P value	0.00176	0.00058	0.65200	0.22200	0.00219
Axial Diffusivity	r	0.21126	0.27346	0.01500	-0.07000	0.46408
	p value	0.10822	0.03450	0.90500	0.58900	0.00015
Radial Diffusivity	r	0.46275	0.50047	-0.03100	-0.23000	0.23155
	p value	0.00015	0.00003	0.81700	0.07400	0.07017

As shown in Table 3, ANCOVA revealed that fractional anisotropy was significantly higher in males than females only for the right ST (p = 0.001, F = 12.71). The value for partial eta² was 0.175, indicative of a moderate effect size. Also, in the right ST the mean and radial diffusivity had initially appeared to be higher in females than in males at a significance of p < 0.05. These findings, however, collapsed after the Bonferroni correction. The same was true for radial diffusivity in the left ST and the left VAP i.e. initially appearing to be significantly different but not surviving Bonferroni correction. None of the other tracts showed any statistically significant relationships between the investigated diffusion metrics and age or gender.

Table 3

ANCOVA comparing differences between male and female subjects for the left and right stria terminalis, the left and right ventral amygdalofugal pathway and the anterior commissure. Corrections made for age and estimated total intracranial volume. Statistically significant values that survived the Bonferroni correction (p < 0.0033) are highlighted in bold. ANCOVA = Analysis of Covariance
	Stria Terminalis
	Left			Right
	F	p	Partial eta2	F	p		Partial eta2
Fractional Anisotropy	5.054	0.028	0.079	12.71	0.001 (M > F)	0.175
Mean Diffusivity	2.669	0.108	0.043	5.653	0.021 (F > M)		0.087
Axial Diffusivity	0.465	0.498	0.008	1.33	0.253		0.022
Radial Diffusivity	3.204	0.078	0.051	8.936	0.004 (F > M)		0.13
	Ventral Amygdalofugal Pathway
	Left			Right
	F	p	Partial eta2	F	p		Partial eta2
Fractional Anisotropy	1.908	0.172	0.031	0.055	0.815		0.001
Mean Diffusivity	1.5	0.225	0.024	0.403	0.528		0.007
Axial Diffusivity	0.293	0.59	0.005	1.33	0.253		0.022
Radial Diffusivity	5.089	0.028	0.083	0.216	0.644		0.004
	Anterior Commissure
	F		p		Partial eta2
Fractional Anisotropy	0.18		0.673		0.003
Mean Diffusivity	0.209		0.649		0.003
Axial Diffusivity	0.046		0.831		0.001
Radial Diffusivity	0.244		0.623		0.004

This is the first study to investigate all three output tracts from the amygdala. This is also the first study to use an anatomically driven tractography protocol to separate the fornix from the ST fibres and also to reconstruct the VAP and three of its branches (to the basal forebrain, hypothalamus and nucleus accumbens).

This study has found that aging affects the diffusion characteristics of these tracts. Few differences in diffusion metrics were observed between male and female subjects; the only exception being the right ST showing a higher fractional anisotropy in males than in females. The inverse relationship observed between age and diffusion metrics for the left ST, right ST and the AC were to be expected; given the general trend of a decrease in white matter integrity as age increases for various white matter tracts throughout the brain (Adalbert & Coleman, 2013). No significant correlation was observed between diffusion metrics and age for the VAP may be explained by the more direct path that the VAP takes in comparison to the C-shaped path of the ST. Moreover, there is a greater proportion of myelinated axons in the VAP than in the ST. These factors may contribute towards a resistance to degenerative changes with age.

Only the right ST revealed a difference in diffusion metrics with sex (increased fractional anisotropy in males). Research has shown that sex differences in brain structure are evident not only in grey matter but also in white matter microstructure, reflecting diverse patterns of neural organisation between males and females (Angelopoulou et al., 2020). A meta-analysis of sex differences in human brain structure revealed that males tend to exhibit larger volumes and higher tissue densities in the left amygdala compared to females (Ruigrok et al., 2014). Studies on white matter microstructure suggest a complex effect of sex on diffusion metrics. Men typically exhibit higher fractional anisotropy in the cerebellar white matter and left superior longitudinal fasciculus, suggesting enhanced motor development and language lateralization, respectively (Ingalhalikar et al., 2014). Women demonstrate higher fractional anisotropy in the corpus callosum, potentially indicating greater interhemispheric connectivity and efficiency (Kanaan et al., 2014).

Of all studies on the output tracts of the amygdala to date, this study has conducted the most complete analysis of these tracts by using the largest sample size, including all amygdala output tracts and analysing the diffusion characteristics of these tracts. A 2016 paper was the first to reveal the VAP using neuroimaging. However, this study only emphasises the tract isolation procedure using probabilistic seeds, without any anatomical landmarks or quantitative analysis of the isolated tracts being performed (Kamali et al., 2016). Similarly, Cohen et al. used diffusion-weighted imaging, in 2008, to depict anatomical information about amygdalar brain circuits in order to study feedback-guided learning (Cohen, Elger, & Weber, 2008). In 2011, Bach et al. also used diffusion-weighted imaging with probabilistic tractography to delineate amygdala connectivity (Bach, Behrens, Garrido, Weiskopf, & Dolan, 2011). Moreover in 2019, Goetschius et al. studied amygdalo-prefrontal cortex white matter tractography in adolescents using probabilistic mapping to gain insights into emotional circuits (Goetschius et al., 2019). These previous studies solely highlight the general connectivity of amygdalar output tracts in their participant rather than relying on robust anatomical landmarks. The research presented in this paper is also the first to find statistically significant correlations of the diffusion metrics of these tracts with age and gender.

It is almost impossible to distinguish the ST from the fornix (Fig. 4) in standard clinical T1 or T2-weighted MRI scans. Two studies to date have attempted to use diffusion-weighted imaging for visualisation and isolation of this white matter tract. One of these studies demonstrated a pre and post-commissural connectivity similar to that of the fornix (Arash Kamali and David, 2015) while the other, using the bed nucleus of the stria terminalis (BNST) as a seed region, showed connectivity of the BNST to the amygdala via the stria terminalis and ansa peduncularis (Kruger, Shiozawa, Kreifelts, Scheffler, & Ethofer, 2015). Both of these studies used probabilistic tractography. While useful for exploratory studies, probabilistic tractography is not anatomically driven and therefore, is not suitable for precise stereotactic localisation which is required to separate out the fornix and stria terminalis along their length. This study’s’ method of using a computer-generated mask of the amygdala itself as an AND gate allowed us to dissect the ST more accurately and relatively cleanly from the arc of the fornix.

The study's findings are subject to several limitations. Firstly, the small sample size of 64 subjects may restrict the generalisability of the results to broader populations. Secondly, variability in the methodological approach, particularly in the placement of ROIs and gates for tractography, could impact the accuracy and reproducibility of the findings. Additionally, despite the use of advanced tractography techniques, the accuracy of such methods in analysing white matter tracts, particularly in anatomically complex regions like the amygdala, may be compromised. Reliance on certain assumptions and parameters in tractography and statistical analyses may introduce biases or overlook important nuances in the data. Further research addressing these limitations could enhance the understanding of gender differences in brain structure

The AC, the ST and the VAP (and 3 of its branches) were successfully isolated in all 64 subjects using high-resolution diffusion tensor imaging. Correlations were observed between increasing age and decreasing fractional anisotropy, mean diffusivity, and radial diffusivity for the left and right ST. Furthermore, a negative correlation of age with mean diffusivity and axial diffusivity for the AC was revealed. No relationship between age and diffusion metrics was noted for the left and right VAP. These findings are in keeping with a general decrease in white matter integrity with age across various white matter tracts in the brain. Upon investigating the relationship of sex on diffusion properties, the only relationship observed was that of the right ST showing a higher fractional anisotropy in males than in females. This is the first time all three tracts have been isolated with measurements of diffusion metrics for sex and age. The novel and anatomically driven protocols presented here for reconstructing the ST and VAP (and the basal forebrain, hypothalamic and nucleus accumbens branch) may be of benefit in future studies investigating the amygdala connectivity and function using diffusion MRI.

Acknowledgements

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by [Anurag Nasa], [Thomas Drago], [Ashka Shah], [Bharti Kewlan], [Katharina Nagassima], [Muhammad Mahmoud] and [Elena Roman]. Tractography performed by [Michael O’Connor], [Emma O’Hora], [Linda Kelly], [Jin zhe Ang], [Arunava Guha], [Arunava Guha], [Michael Connaughton], [Orla Mitchell]. Lead investigators were [Kirk Levins], [Eric O’Hanlon], [Veronica O’Keane] and [Darren William Roddy] The first draft of the manuscript was written by [Anurag Nasa], [Thomas Drago] and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

The datasets generated during and/or analysed during the current study are part of the the REDEEM (Research in Depression: Endocrinology, Epigenetics, and neuroiMaging) research study at Trinity College Dublin. Data is available from the authors upon reasonable request and with permission from Trinity College Dublin.

Author Contribution

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by [A.N], [T.D], [A.S], [B.K], [K.N], [M.M] and [E.R]. Tractography performed by [M.O.C], [E.O.H], [L.K], [J.G], [A.G], [A.G], [M.C], [O.M]. Lead investigators were [K.L], [Eric O’Hanlon], [V.O.K] and [D.W.R] The first draft of the manuscript was written by [A.N], [T.D] and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

AbuHasan Q, Reddy V, Siddiqui W (2020) Neuroanatomy, Amygdala. In
Adalbert R, Coleman M (2013) Axon pathology in age-related neurodegenerative disorders. Neuropathol Appl Neurobiol 39(2):90–108
Alheid GF (2003) Extended amygdala and basal forebrain. Ann N Y Acad Sci 985(1):185–205
Amaral DG (2003) The amygdala, social behavior, and danger detection. Ann N Y Acad Sci 1000(1):337–347
Amunts K, Kedo O, Kindler M, Pieperhoff P, Mohlberg H, Shah NJ, Zilles K (2005) Cytoarchitectonic mapping of the human amygdala, hippocampal region and entorhinal cortex: intersubject variability and probability maps. Anat Embryol (Berl) 210(5–6):343–352. 10.1007/s00429-005-0025-5
Kamali A, David MY, a. D. D. L. a. H. I. S. a. S. P. J. a. Z. K. a. R. F. R. a. K. M. (2015). Mapping the trajectory of the stria terminalis of the human limbic system using high spatial resolution diffusion tensor tractography. Neurosci Lett, 608, 45–50. https://doi.org/10.1016/j.neulet.2015.09.035
Bach DR, Behrens TE, Garrido L, Weiskopf N, Dolan RJ (2011) Deep and superficial amygdala nuclei projections revealed in vivo by probabilistic tractography. J Neurosci 31(2):618–623
Bao AM, Swaab DF (2019) The human hypothalamus in mood disorders: The HPA axis in the center. IBRO Rep 6:45–53. 10.1016/j.ibror.2018.11.008
Basser PJ, Pajevic S, Pierpaoli C, Duda J, Aldroubi A (2000) In vivo fiber tractography using DT-MRI data. Magn Reson Med 44(4):625–632. 10.1002/1522-2594(200010)44:4<625::aid-mrm17>3.0.co;2-o
Baydin S, Gungor A, Tanriover N, Baran O, Middlebrooks EH, Rhoton AL Jr (2017) Fiber tracts of the medial and inferior surfaces of the cerebrum. World Neurosurg 98:34–49
Berridge KC, Kringelbach ML (2015) Pleasure systems in the brain. Neuron 86(3):646–664. 10.1016/j.neuron.2015.02.018
Cassell MD, Freedman LJ, Shi C (1999) The intrinsic organization of the central extended amygdala. Ann N Y Acad Sci 877(1):217–241
Cohen MX, Elger CE, Weber B (2008) Amygdala tractography predicts functional connectivity and learning during feedback-guided decision-making. NeuroImage 39(3):1396–1407
Davis M (1994) The role of the amygdala in emotional learning. International review of neurobiology, vol 36. Elsevier, pp 225–266
Davis M, Shi C (1999) The extended amygdala: are the central nucleus of the amygdala and the bed nucleus of the stria terminalis differentially involved in fear versus anxiety? Ann N Y Acad Sci 877:281–291. 10.1111/j.1749-6632.1999.tb09273.x
Duvarci S, Pare D (2014) Amygdala microcircuits controlling learned fear. Neuron 82(5):966–980
Farquharson S, Tournier JD, Calamante F, Fabinyi G, Schneider-Kolsky M, Jackson GD, Connelly A (2013) White matter fiber tractography: why we need to move beyond DTI. J Neurosurg 118(6):1367–1377. 10.3171/2013.2.jns121294
Fischl B (2012) FreeSurfer. NeuroImage 62(2):774–781. 10.1016/j.neuroimage.2012.01.021
Gaughan C, Nasa A, Roman E, Cullinane D, Kelly L, Riaz S, Mosley O (2023) A Pilot Study of Adolescents with Psychotic Experiences: Potential Cerebellar Circuitry Disruption Early Along the Psychosis Spectrum. Cerebellum, 1–11
Goetschius LG, Hein TC, Mattson WI, Lopez-Duran N, Dotterer HL, Welsh RC, Monk CS (2019) Amygdala-prefrontal cortex white matter tracts are widespread, variable and implicated in amygdala modulation in adolescents. NeuroImage 191:278–291
Gray TS (1999) Functional and anatomical relationships among the amygdala, basal forebrain, ventral striatum, and cortex: an integrative discussion. Ann N Y Acad Sci 877(1):439–444
Haller J (2018) The role of central and medial amygdala in normal and abnormal aggression: A review of classical approaches. Neurosci Biobehavioral Reviews 85:34–43
Hamilton M (1959) The assessment of anxiety states by rating. Br J Med Psychol 32:50–55. 10.1111/j.2044-8341.1959.tb00467.x
Hamilton M (1960) A rating scale for depression. J Neurol Neurosurg Psychiatry 23(1):56–62. 10.1136/jnnp.23.1.56
Hsu T-T, Huang T-N, Hsueh Y-P (2020) Anterior Commissure Regulates Neuronal Activity of Amygdalae and Influences Locomotor Activity, Social Interaction and Fear Memory in Mice. Front Mol Neurosci 13(47). 10.3389/fnmol.2020.00047
IBM Corp Released 2019. IBM SPSS Statistics for Windows, Version 26.0. Armonk, NY: IBM Corp. In
Irfanoglu MO, Sarlls J, Nayak A, Pierpaoli C (2019) Evaluating corrections for Eddy-currents and other EPI distortions in diffusion MRI: methodology and a dataset for benchmarking. Magn Reson Med 81(4):2774–2787. https://doi.org/10.1002/mrm.27577
Jeurissen B, Leemans A, Jones DK, Tournier JD, Sijbers J (2011) Probabilistic fiber tracking using the residual bootstrap with constrained spherical deconvolution. Hum Brain Mapp 32(3):461–479. 10.1002/hbm.21032
Johansen-Berg H, Rushworth MF (2009) Using diffusion imaging to study human connectional anatomy. Annu Rev Neurosci 32:75–94. 10.1146/annurev.neuro.051508.135735
Jones DK, Knösche TR, Turner R (2013) White matter integrity, fiber count, and other fallacies: the do's and don'ts of diffusion MRI. NeuroImage 73:239–254
Jones DK, Leemans A (2011) Diffusion tensor imaging. Magnetic resonance neuroimaging. Springer, pp 127–144
Kamali A, Sair HI, Blitz AM, Riascos RF, Mirbagheri S, Keser Z, Hasan KM (2016) Revealing the ventral amygdalofugal pathway of the human limbic system using high spatial resolution diffusion tensor tractography. Brain Struct Funct 221(7):3561–3569. 10.1007/s00429-015-1119-3
Kamali A, Yousem DM, Lin DD, Sair HI, Jasti SP, Keser Z, Hasan KM (2015) Mapping the trajectory of the stria terminalis of the human limbic system using high spatial resolution diffusion tensor tractography. Neurosci Lett 608:45–50. 10.1016/j.neulet.2015.09.035
Kiernan J (2012) Anatomy of the temporal lobe. Epilepsy research and treatment, 2012
Koller K, Hatton CM, Rogers RD, Rafal RD (2019) Stria terminalis microstructure in humans predicts variability in orienting towards threat. Eur J Neurosci 50(11):3804–3813. https://doi.org/10.1111/ejn.14504
Kruger O, Shiozawa T, Kreifelts B, Scheffler K, Ethofer T (2015) Three distinct fiber pathways of the bed nucleus of the stria terminalis to the amygdala and prefrontal cortex. Cortex 66:60–68. 10.1016/j.cortex.2015.02.007
Le Bihan D, Mangin JF, Poupon C, Clark CA, Pappata S, Molko N, Chabriat H (2001) Diffusion tensor imaging: concepts and applications. J Magn Reson Imaging 13(4):534–546
Lee Y, Davis M (1997) Role of the hippocampus, the bed nucleus of the stria terminalis, and the amygdala in the excitatory effect of corticotropin-releasing hormone on the acoustic startle reflex. J Neurosci 17(16):6434–6446. 10.1523/jneurosci.17-16-06434.1997
Leemans A, Jeurissen B, Sijbers J, Jones D (2009) ExploreDTI: a graphical toolbox for processing, analyzing, and visualizing diffusion MR data
Mai JK, Majtanik M, Paxinos G (2015) Atlas of the human brain. Academic
Miller EJ, Marie S, Breier LR, M. R., Swerdlow NR (2010) Pathways from the ventral hippocampus and caudal amygdala to forebrain regions that regulate sensorimotor gating in the rat. Neuroscience 165(2):601–611. 10.1016/j.neuroscience.2009.10.036
Mori S, Kageyama Y, Hou Z, Aggarwal M, Patel J, Brown T, Troncoso JC (2017) Elucidation of White Matter Tracts of the Human Amygdala by Detailed Comparison between High-Resolution Postmortem Magnetic Resonance Imaging and Histology. Front Neuroanat 11:16. 10.3389/fnana.2017.00016
Nasa A, Gaughan C, Mahmoud M, Kelly JR, Roman E, Levins KJ, O'Keane V (2021) The human dorsal hippocampal commissure: Delineating connections across the midline using multi-modal neuroimaging in major depressive disorder. Neuroimage: Rep 1(4):100062
Nave KA, Werner HB (2014) Myelination of the nervous system: mechanisms and functions. Annu Rev Cell Dev Biol 30:503–533. 10.1146/annurev-cellbio-100913-013101
Noback CR, Ruggiero DA, Demarest RJ, Strominger NL (2005) The human nervous system: structure and function. Springer Science & Business Media
Nolan M, Roman E, Nasa A, Levins KJ, O’Hanlon E, O’Keane V, Roddy W, D (2020) Hippocampal and Amygdalar Volume Changes in Major Depressive Disorder: A Targeted Review and Focus on Stress. Chronic Stress 4:2470547020944553
O'Neill A, Dooley N, Healy C, Carey E, Roddy D, Frodl T, Cannon M (2022) Longitudinal grey matter development associated with psychotic experiences in young people. Biol Psychiatry Global Open Sci
O’Neill A, Dooley N, Roddy D, Healy C, Carey E, Frodl T, Cannon M (2024) Longitudinal hippocampal subfield development associated with psychotic experiences in young people. Translational psychiatry 14(1):44
Pascalau R, Popa Stanila R, Sfrangeu S, Szabo B (2018) Anatomy of the Limbic White Matter Tracts as Revealed by Fiber Dissection and Tractography. World Neurosurg 113:e672–e689. 10.1016/j.wneu.2018.02.121
Perrone D, Aelterman J, Pizurica A, Jeurissen B, Philips W, Leemans A (2015) The effect of Gibbs ringing artifacts on measures derived from diffusion MRI. NeuroImage 120:441–455. 10.1016/j.neuroimage.2015.06.068
Porrino L, Crane A, Goldman-Rakic P (1981) Direct and indirect pathways from the amygdala to the frontal lobe in rhesus monkeys. J Comp Neurol 198(1):121–136
Roddy D, Kelly JR, Farrell C, Doolin K, Roman E, Nasa A (2021). . O'Hanlon, E. Amygdala substructure volumes in Major Depressive Disorder. NeuroImage: Clinical, 102781
Roddy DW, Farrell C, Doolin K, Roman E, Tozzi L, Frodl T, O’Hanlon E (2018) The Hippocampus in Depression: More Than the Sum of Its Parts? Advanced Hippocampal Substructure Segmentation in Depression. Biological Psychiatry
Roddy DW, Roman E, Nasa A, Gazzaz A, Zainy A, Burke T, Clarke M (2022) Microstructural changes along the cingulum in young adolescents with psychotic experiences: an along-tract analysis. Eur J Neurosci
Soares J, Marques P, Alves V, Sousa N (2013) A hitchhiker's guide to diffusion tensor imaging. Front NeuroSci 7(31). 10.3389/fnins.2013.00031
Soares JM, Marques P, Alves V, Sousa N (2013) A hitchhiker's guide to diffusion tensor imaging. Front Neurosci 7:31. 10.3389/fnins.2013.00031
Stamatakis AM, Sparta DR, Jennings JH, McElligott ZA, Decot H, Stuber GD (2014) Amygdala and bed nucleus of the stria terminalis circuitry: implications for addiction-related behaviors. Neuropharmacology 76:320–328
Tournier JD, Mori S, Leemans A (2011) Diffusion tensor imaging and beyond. Magn Reson Med 65(6):1532–1556. 10.1002/mrm.22924
Vos SB, Tax CMW, Luijten PR, Ourselin S, Leemans A, Froeling M (2017) The importance of correcting for signal drift in diffusion MRI. Magn Reson Med 77(1):285–299. https://doi.org/10.1002/mrm.26124
Winter TJ, Franz EA (2014) Implication of the anterior commissure in the allocation of attention to action. Front Psychol 5:432

No competing interests reported.

Download PDF

Editorial decision: Revision requested
30 Jul, 2024
Reviews received at journal
19 Jul, 2024
Reviews received at journal
05 Jul, 2024
Reviewers agreed at journal
04 Jul, 2024
Reviewers agreed at journal
04 Jul, 2024
Reviewers invited by journal
03 Jul, 2024
Editor assigned by journal
26 Jun, 2024
Submission checks completed at journal
22 Jun, 2024
First submitted to journal
20 Jun, 2024

You are reading this latest preprint version

Imaging the three human primary amygdala output tracts with age and sex characterisation across the lifespan

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Introduction

Methods and Materials

Participants

MRI Data Acquisition

Image Analyses Using FreeSurfer 6.0

DWI Pre-processing and Whole Brain Tractography

Tract isolation

Cleaning and Segmentation

Statistical Analysis

Results

Discussion

Conclusion

Declarations

Author Contribution

References

Additional Declarations

Status:

Version 1