The prevalence of obesity and associated health conditions is increasing in the developed world. Obesity is related to atrophy and dysfunction of the hippocampus and hippocampal lesions may lead to increased appetite and weight gain. The hippocampus is connected via the fornix tract to the hypothalamus, orbitofrontal cortex, and the nucleus accumbens, all key structures for homeostatic and reward related control of food intake. The present study employed diffusion MRI tractography to investigate the relationship between microstructural properties of the fornix and variation in Body Mass Index (BMI), within normal and overweight ranges, in a group of community-dwelling older adults (53–93 years old). Larger BMI was associated with larger axial and mean diffusivity in the fornix (r = 0.64 and r = 0.55 respectively), relationships that were most pronounced in overweight individuals. Moreover, controlling for age, education, cognitive performance, blood pressure and global brain volume increased these correlations. Similar associations were not found in the parahippocampal cingulum, a comparison temporal association pathway. Thus, microstructural changes in fornix white matter were observed in older adults with increasing BMI levels from within normal to overweight ranges, so are not exclusively related to obesity. We propose that hippocampal-hypothalamic-prefrontal interactions, mediated by the fornix, contribute to the healthy functioning of networks involved in food intake control. The fornix, in turn, may display alterations in microstructure that reflect weight gain.
Citation: Metzler-Baddeley C, Baddeley RJ, Jones DK, Aggleton JP, O’Sullivan MJ (2013) Individual Differences in Fornix Microstructure and Body Mass Index. PLoS ONE 8(3): e59849. doi:10.1371/journal.pone.0059849
Editor: Carles Soriano-Mas, Bellvitge Biomedical Research Institute-IDIBELL, Spain
Received: November 24, 2012; Accepted: February 21, 2013; Published: March 28, 2013
Copyright: © 2013 Metzler-Baddeley et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This research was funded by the Medical Research Council, UK, via a Clinician Scientist Fellowship to MOS (G0701912). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The prevalence of obesity and its associated health problems is growing in the developed world. To develop effective interventions for obesity prevention, it is important to understand the neural and psychological factors that contribute to chronic weight gain. In the current study we investigated in a group of community dwelling, high functioning older adults the relationship between individual differences in Body Mass Index (BMI) and variation in the white matter microstructure of the fornix, the principal fibre tract associated with hippocampal connections beyond the temporal lobe .
Research into hippocampal function has traditionally focused on learning and memory  but there is growing appreciation of its involvement in other functions  including the regulation of food intake. Hippocampal lesions in rats can result in increased appetitive behaviour and weight gain , . Likewise, reduced hippocampal volume  and abnormal hippocampal activation in response to food stimulation or feeding manipulations  have been related to obesity, although the direction of causality remains unclear.
The hippocampus is known to be directly connected, via the fornix, with several regions involved in the control of food intake. One such target is the hypothalamus, a complex region involved in homeostatic functions . Hippocampal projections reach the preoptic area, anterior hypothalamic nuclei, arcuate nucleus, ventromedial nucleus and lateral hypothalamic area . The precommissural fornix also contains efferents to nucleus accumbens and the orbital frontal cortices , . These connections innervate sites strongly implicated in the hedonic control of food intake , .
Diffusion MRI tractography ,  was used to investigate the relationship between BMI and microstructure of the fornix, as well as a comparison tract, the parahippocampal cingulum, which also contains medial temporal lobe connections but is not directly linked to either medial prefrontal cortex or hypothalamus , . Microstructural indices were derived for each tract. The fornix is a medium-sized pathway that runs through the ventricles and, hence, is particularly susceptible to cerebrospinal fluid (CSF) based partial volume artefacts in diffusion tensor imaging (DTI) indices . This is especially problematic in older adults, as aging is associated with significant gray and white matter atrophy , . To account for this confound, data were corrected with the Free Water Elimination (FWE) approach . This method fits two tensors, an isotropic and an anisotropic compartment, to the diffusion data from each voxel and, hence, allows the elimination of CSF contamination in white matter microstructural indices.
Additional measurements were made of hippocampal and whole brain volume, and associations were controlled for potential confounds of volume measures, age, sex, education, cognitive function (verbal IQ and episodic memory), blood pressure and diabetes status.
This study was reviewed and approved by the South East Wales Research Ethics Committee and each participant provided informed, written consent. All investigations have been conducted according to the principles expressed in the Declaration of Helsinki.
Forty-six participants over the age of 50 years were recruited for a project investigating aging and brain structure from the local community by advertisements posted on the internet, and in family physician waiting rooms, newsletters and mail shots. Participants were informed that the study was about aging and memory; BMI and risk factors were not mentioned explicitly so are unlikely to have influenced recruitment. Participants fulfilled the selection criteria for the study if they were free of a history of neurological and/or psychiatric disease, moderate or severe head injury, alcohol and/or drug abuse, memory or other cognitive decline, stroke, large artery or peripheral vascular disease, structural heart disease or heart failure, and contra-indications to MRI. Two participants withdrew from the study due to ill health and one participant was excluded due to a subsequent diagnosis of Parkinson’s Disease. After visual inspection of the structural MRI scans for overt pathology three further participants were excluded because of extensive white matter hyperintensities suggestive of cerebral microvascular disease . One individual’s MRI data could not be analysed due to motion artefacts. Finally, one participant’s height was not recorded at the beginning of the study leaving 38 MRI and BMI datasets available for analysis in the present study (see Table 1).
Risk factors for cardiovascular disease such as increased Body Mass Index (BMI) (weight in kg/height in square meter), hypertension, diabetes mellitus, high levels of blood cholesterol, and smoking were documented. Each participant was weighed in their clothing without shoes on a mechanical scale. Standing height was measured using a tape measure placed in fixed position against the wall. Systolic and diastolic blood pressure and pulse were measured with a digital blood pressure monitor (Model UA-631; A&D Medical, Tokyo, Japan) whilst participants were relaxed and comfortably seated with their arm well supported on a table. Other cardiovascular risk factors were self-reported by participants in a comprehensive medical history questionnaire.
For 17 participants BMI fell within normal range (18.5–24.9), for 17 within overweight range (25–30), three participants were obese (BMI >30), and one individual was underweight (BMI <18.5). Sixteen participants reported medication to control hypertension, two reported statins to control cholesterol levels, and one person was on medication for Type II diabetes mellitus. Seventeen participants reported never to have smoked and twenty-one participants to have smoked in the past. The group performed at superior level of intelligence [Mean IQ = 121; SD = 8 in the National Adult Reading Test (NART)]  and within normal-superior range for their age in a number of memory and executive function tasks (Table 1 and for detailed description see ). In addition, participants were screened for depression with the 15 items Geriatric Depression Scale (GDS15/. All participants scored within the normal range except for one individual whose score was on the conventional cut-off of 6.
MRI Data Acquisition
Diffusion weighted MR data were acquired using a 3T GE HDx MRI system (General Electric Healthcare) with a twice-refocused spin-echo EPI sequence providing whole oblique axial (parallel to the commissural plane) brain coverage. Data acquisition was peripherally gated to the cardiac cycle. Data were acquired from 60 slices of 2.4 mm thickness, with a field of view of 23 cm, and an acquisition matrix of 96×96. TE was 87 ms and parallel imaging (ASSET factor = 2) was employed. The b-value was 1200 s/mm−2. In each imaging session, data were acquired with diffusion encoded along 30 isotropically distributed directions  and three non-diffusion weighted scans using an optimised gradient vector scheme. The acquisition time was approximately 13 minutes.
Diffusion MRI Pre-processing
The acquired images were corrected for distortions, introduced by the diffusion-weighting gradients, and for motion artefacts using ExploreDTI . This was achieved using a global affine registration of each image volume to the first non-diffusion weighted volume, using normalized mutual information as the cost-function, followed by appropriate re-orientation of the encoding vectors and modulation of the signal intensity by the Jacobian determinant of the transformation . Two separate processing steps were then taken: the first was to fit a two compartment model using the FWE approach to the data, to yield partial volume corrected indices of fractional anisotropy (FA), mean diffusivity (MD), radial diffusivity (RD) and axial diffusivity (AD) ; the second was to use constrained spherical deconvolution (CSD) to extract, in each voxel, the fibre orientation density function (fODF) with the spherical harmonics order no higher than lmax = 6 .
Diffusion MRI Tractography
Deterministic tractography was performed using ExploreDTI , . The tracking algorithm estimated the peak in the fODF at each seed-point and propagated in 0.5 mm steps along this axis. Peaks in the fODF were then estimated at the new location and the tracking moved a further 0.5 mm along the axis subtending the smallest angle to the current trajectory. In this way a pathway was traced through the data until the fODF fell below an arbitrary threshold (in this case 0.1) or the direction of the pathway changed through an angle greater than 60°. The procedure was then repeated by tracking in the opposite direction back to the initial seed-point to reconstruct the whole pathway passing through the seed-point. Three dimensional fibre reconstructions were made by: (i) initial tracking of pathways in the whole brain using every voxel as a seed-point from which to initiate the tracking algorithm; (ii) the subsequent application of waypoint region of interest (ROI) gates (“AND”, “OR”, “NOT” gates following Boolean logic) to isolate specific tracts from the whole brain tractography data. ROIs were manually drawn on a colour coded fibre orientation map  for each individual dataset as described in  using landmark techniques that have previously been shown to be highly reproducible , .
A coronal OR ROI was placed around the anterior pillars and an axial AND ROI around the crus fornici. NOT ROIs were placed rostral to the anterior pillars, caudal to the crus fornici, around the corpus callosum and the upper pons (see Figure 1a).
Seed (OR) waypoint region of interest gates are illustrated in blue, AND gates in green and NOT gates in red.
One axial AND ROI was placed dorsal to the ventral limit of the splenium and another at the level of the pons-midbrain junction . A NOT ROI was placed above the body of the corpus callosum caudal to the rostral-caudal midpoint of the body of the corpus callosum (see Figure 1b).
Hippocampal and Whole Brain Volume Measures
Hippocampal volumes were generated with the HippoQuant software , developed by BioClinica SAS, Lyon, France. This is an adaptation of hippocampal outlining techniques (based on ) but involves placing approximately 80 landmark points, rather than tracing full outlines on each slice, with the remaining surface filled in by deforming a standard three-dimensional hippocampal template to match these landmark points. In a first step, T1-weighted images were displayed simultaneously in all three planes (axial, sagittal, coronal) and a trained rater placed multiple landmarks on the boundary of the hippocampus, paying particular attention to areas of difficult anatomical distinction (for example, the alveus and the boundary with the amygdala). The hippocampal measurements included the hippocampus proper (CA-1 to CA-4 sectors), dentate gyrus, alveus, fimbria and subiculum. Parahippocampal regions (entorhinal cortex and intervening white matter including perforant pathway) were excluded. The resulting hippocampal mask was further refined by a tissue-type segmentation step (i.e. voxels segmented as grey or white matter are included, CSF voxels excluded). The final hippocampal masks were then checked visually by the rater and edited for obvious errors. Hippocampal volumes were corrected for head size by calculating intracranial volume from summing segmented grey, white matter and CSF images (from the volumetric T1-weighted image) in Statistical Parametric Mapping 8 . Brain parenchymal fraction (BPF), i.e. the volume of gray and white matter normalised to total intracranial volume to correct for head size, as a measure for whole brain volume was also obtained.
All participants underwent neuropsychological assessment of working and episodic memory and executive function. Age-related changes for this sample were present only in episodic memory performance and were strongest for verbal free recall in the Free and Cued Selective Reminding task . Performance in this task was also related to variation in fornix microstructure (FA) . Changes in episodic memory in older age may, in some cases, be indicative of underlying neurodegenerative disease such as Alzheimer’s Disease and the prodromal manifestation of Mild Cognitive Impairment (MCI). MCI and Alzheimer’s disease are both associated with atrophy of the medial temporal lobe, including the hippocampus and fornix, and changes in BMI due to weight loss . For these reasons, we employed verbal free recall performance in addition to verbal NART-IQ (see Table 1) as measures of individual variation in cognitive performance that would potentially confound any relationship between fornix microstructure and BMI, if covert Alzheimer’s neuropathology were driving these associations.
Statistical analyses were performed using Predictive Analytics Software (PASW) Statistics 18.0 . BMI data were tested for the assumption of normal distribution with the Kolmogorov-Smirnov and Shapiro-Wilk tests. Two-tailed Pearson product moment correlations were calculated between BMI and white matter microstructural indices in the fornix and the parahippocampal cingulum, as well as with total hippocampal volume and brain parenchymal fraction. These correlations were carried out for all participants (n = 38) and only for individuals with BMI within normal-overweight range (n = 34) to ensure that correlations were not driven by a few extreme cases. Correlations between BMI and structural indices (10 in total) had to reach p<0.005 to comply with experiment-wise Bonferroni corrected significance level of 0.05.
Relationships were further explored separately for normal and overweight participants and females and males. The latter correlations were evaluated with one-tailed tests and were not further subjected to multiple-comparison corrections.
Pearson product moment correlations between white matter microstructural indices in the fornix and potentially confounding variables of age, education, NART-IQ, verbal free recall, systolic and diastolic blood pressure and BPF were calculated. Correlations (a total of 28 comparisons) had to reach p<0.002 to comply with experiment-wise Bonferroni corrected significance level of 0.05.
The role of potential confounds in the relationship between BMI and white matter microstructural indices were explored with partial correlations and multivariate hierarchical regression analyses, which were not further subjected to multiple comparison corrections. Multivariate hierarchical regression analysis was employed by entering all confounding variables first into a linear regression model with BMI as dependent variable, followed by the stepwise addition of the diffusion MRI indices in the fornix and the parahippocampal cingulum. Only one participant in the current study was diabetic – the exclusion of this individual did not alter any associations.
Relationship between BMI and White Matter Microstructure
Both Kolmogorov-Smirnov and Shapiro-Wilk tests of normality produced non-significant results for the BMI data consistent with the null hypothesis of normality (F(38) = 0.08, p = 0.20; W(38) = 0.98, p = 0.66).
For all 38 participants there were positive correlations between BMI and fornix AD (r = 0.49; p≤0.002) and BMI and fornix MD (r = 0.46; p≤0.004) as well as a trend for a positive correlation between BMI and fornix RD (r = 0.37; p = 0.02). No correlations were observed between BMI and fornix FA (r = 0.08, p = 0.64) or any microstructural indices in the parahippocampal cingulum (FA: r = 0.20, p = 0.23; MD: r = −0.09, p = 0.61; RD: r = −0.14, p = 0.42; AD: r = 0.07, p = 0.68).
After the exclusion of the four extreme cases (three obese and one underweight individual) (n = 34) the positive correlations between BMI and axial and mean diffusivity in the fornix increased in magnitude (AD: r = 0.64, p≤0.001; MD: r = 0.55; p≤0.001) (see Table 2, Figure 2). A clearer trend for a relationship between BMI and fornix RD was observed (r = 0.45, p = 0.007), but again no correlations were found with fornix FA (r = 0.18, p = 0.32) or with any of the microstructural indices in the parahippocampal cingulum (FA: r = 0.13, p = 0.46; MD: r = 0.07, p = 0.69; AD: r = 0.14, p = 0.42; RD: r = 0.14, p = 0.42) (see Table 2, Figure 2).
R2 refers to the correlation coefficients based on the normal and overweight sample only (n = 34). BMI ranges are color coded with blue indicating underweight, green healthy, orange overweight and red obese ranges.
Further exploratory analyses (one-tailed) revealed that participants with normal weight (n = 17) showed no significant relationships between BMI and microstructural indices in the fornix (FA: r = 0.23, p = 0.19; MD: r = 0.12, p = 0.65; AD: r = 0.22, p = 0.20; RD: r = 0.01, p = 0.49). In contrast, overweight participants (n = 17) exhibited clear positive relationships (uncorrected) between all diffusivity measures in the fornix (MD: r = 0.44, p≤0.04; AD: r = 0.47, p≤0.03; RD: r = 0.42, p≤0.05) but not with fornix FA (r = −0.11, r = 0.34).
Females (n = 22) and males (n = 16) demonstrated comparable positive correlations between BMI and fornix AD (males: r = 0.45, p≤0.03; females: 0.46, p≤0.02) and fornix MD (males: r = 0.48, p≤0.03; females: 0.37, p≤0.04). Both groups exhibited comparable trends between BMI and fornix RD (males: r = 0.36, p = 0.09; females: r = 0.32, p = 0.07) but neither group showed any correlations between BMI and fornix FA (males: r = 0.08, p = 0.39; females: r = 0.12, p = 0.29).
Relationships with Confounding Variables
Since the strongest correlations were found for the combined group of normal and overweight participants (BMI 18.5–30) and the aim of our study was to investigate factors that may contribute to an individual’s susceptibility to gain weight, the following analyses did not include the three obese and the one underweight individual (n = 34).
There were no correlations between BMI and any of the following variables: age (r = −0.04; p = 0.84), systolic and diastolic blood pressure (r = 0.01; p = 0.99 and r = −0.04; p = 0.84), education (r = −0.19; p = 0.28), NART-IQ (r = 0.28; p = 0.11), verbal free recall (r = 0.24; p = 0.17) or BPF (r = 0.17; p = 0.35).
Further, apart from the already reported positive relationships between fornix FA and age and verbal free recall respectively , there were no relationships between fornix microstructure and age, blood pressure, education, or BPF (see Table 3). However, trends (uncorrected for multiple comparisons) were observed between fornix diffusivity indices and NART-IQ as well as free recall (see Table 3).
Partial correlations between BMI and fornix diffusivity indices (AD, MD, RD), accounting for the variation in potentially confounding variables of age, education, NART-IQ, free recall, blood pressure and brain volume did not remove, and in some cases increased, the magnitude of correlations. The partial correlations between BMI and fornix microstructural indices accounting for age were r = 0.64, p≤0.001 (AD); r = 0.56, p≤0.001 (MD); r = 0.49, p≤0.004 (RD); accounting for education: r = 0.68, p≤0.001 (AD); r = 0.58, p≤0.001 (MD); r = 0.50, p≤0.001 (RD); for NART-IQ: r = 0.59, p≤0.001 (AD); r = 0.52, p≤0.002 (MD); r = 0.44, p≤0.01 (RD); accounting for free recall: r = 0.62, p≤0.001 (AD); r = 0.53, p≤0.002 (MD); r = 0.46, p≤0.007 (RD); accounting for blood pressure (systolic and diastolic): r = 0.65, p≤0.001 (AD); r = 0.55, p≤0.001 (MD); r = 0.46, p≤0.01 (RD) and accounting for overall brain volume with BPF: r = 0.64, p≤0.001 (AD); r = 0.59, p≤0.001 (MD); r = 0.52, p≤0.002 (RD).
Age, education, NART-IQ, verbal free recall, systolic and diastolic blood pressure and BPF were entered together into a hierarchical regression model with BMI as dependent variable, followed by the stepwise addition of all microstructural indices (AD, RD, MD, FA) in the fornix and the parahippocampal cingulum.
The adjusted R2 for all confounding variables together (age, education, IQ, free recall, blood pressure, BPF) was 0.23 with no single variable making a significant independent contribution. The inclusion of white matter microstructural indices improved the fit of the model from 23% to 45% (adjusted R2 = 0.45) with fornix AD being the only variable accounting for a significant proportion of the variance in BMI (t = 2.9, p≤0.01) independent of the other variables. The pattern of results was similar when free recall was not used as a covariate. In this case the inclusion of white matter microstructural indices improved the fit from 15% to 46% (adjusted R2 = 0.46) with fornix AD the sole independent predictor (t = 3.5, p≤0.002).
Relationship between Body Weight and Hippocampal Volume
There was no correlation between BMI and total hippocampal volume (r = 0.26; p = 0.14).
Positive relationships between BMI and fornix microstructure were observed in a group of older participants. These relationships were primarily driven by overweight but not obese individuals and were more pronounced after the exclusion of three obese and one underweight participant, so were not determined by outliers. The relationships between fornix microstructural indices and BMI were comparable between females and males, and were not confounded by age, education, IQ, episodic memory performance, high blood pressure, diabetes mellitus or inter-individual variation in whole brain volume. Thus, we propose that the observed correlation in our cognitively high functioning group of older people was not mediated by individual differences in demographic or intellectual variables, or by cardio-vascular risk states such as high blood pressure or diabetes.
The regulation of food intake is a complex process involving interactions between regions responsible for homeostatic, habitual, reward-related and mnemonic aspects of behavioural control. The hypothalamus monitors energy levels and responds to them adaptively through hunger and satiety signals, and is the hub for homeostatic aspects of food regulation. The hippocampus may contribute learning and memory processes that influence the control of appetitive and consummatory behavior, such as experiencing hunger at set meal times during the day , . Other hippocampal interactions via the fornix are with the ventral striatum , a key structure in dopamine-based habitual reward mechanisms, ensuring positive reward associated with food intake. Medial prefrontal and orbitofrontal cortices contribute to food regulation by mediating higher level reward and cognitive control aspects , , important for the successful maintenance of a healthy body weight. In addition, a variety of subcortical sites innervate the hippocampus via the fornix .
All of these regions contribute unique attributes but need to interact with each other to ensure the adaptive and efficient regulation of food intake behaviour. For instance, hippocampal-prefrontal-ventral striatum circuits are involved in inhibitory aspects of response control  and also mediate the ability to delay anticipated reward . Further, hippocampal-hypothalamic connections may facilitate learned patterns of food intake control and energy balance . The fornix maintains projections between hippocampus and prefrontal cortex, ventral striatum, and hypothalamus respectively. Thus, one interpretation of the observed association between fornix microstructure and BMI is that the fornix, in connecting all of these regions, contributes critically to the interactive processes that underlie food intake regulation.
The relationship with BMI was specific to the fornix and was not observed for the parahippocampal cingulum. The parahippocampal cingulum also contains medial temporal lobe connections, but in contrast to the fornix is preferentially linked to occipital and parietal cortices, rather than the medial prefrontal cortex or hypothalamus . Further within the current sample of normal and overweight adults, there was no correlation between BMI and hippocampal volume. Together, this pattern of results suggests that not all hippocampal/medial temporal lobe connections contribute substantially but only those that project directly to regions involved in food regulation such as the hypothalamus and the prefrontal cortex.
One other study  has reported a relationship between fornix microstructure and obesity. There are, however, a number of important differences in the present study. Stanek et al.  investigated white matter changes across the whole brain and found decreases in fornix FA (the authors did not report measures of diffusivity) in obese individuals. Obesity is associated with a number of metabolic and pathological abnormalities, so in this context structural brain changes are less surprising. In contrast, the present study observed changes in fornix diffusivity in individuals with BMI ranges in the normal and overweight range demonstrating that such effects are not exclusively related to obesity and perhaps emerge before the transition to obesity.
It should be noted that the absence of a correlation between variations in BMI and blood pressure and other health conditions may be due to range restrictions in the health variables and the known difficulties of quantifying lifetime risk exposures, especially when they are effectively controlled by medication, as in our sample. In addition, it is unlikely that biological features of depression would account for our results since participants were free of a history of major depressive disorder and – all but one individual, scored within normal range in the GDS15 with even that individual right on the cut-off score.
Alterations in fornix microstructure were specifically observed with axial, mean, and radial diffusivity but were not present for fractional anisotropy. Combined with our previous work on age-related changes in fornix microstructure and episodic memory in the same sample of older adults , we observed a double dissociation: partial volume corrected FA was significantly related with age and memory but not with BMI; whereas diffusivity indices were related to BMI but not to age or memory performance. This pattern of results suggests that FA and diffusivity indices may be sensitive to different microstructural aspects of white matter and may not always be interchangeable or directly comparable.
It is difficult to infer the precise biophysical alterations underlying changes in DTI based indices of FA, MD, AD and RD because these measures can be modulated by a number of parameters such as myelination, axonal density and diameter and the complexity of the underlying fibre architecture . Interpretations of DTI based diffusivity indices in terms of axonal loss or myelin alterations are primarily based on results from animal studies that have investigated single coherently aligned fibres . Since the fibre architecture in the human brain, however, is complex and scenarios such as crossing fibres affect the magnitude of DTI based indices (FA for instance drops in areas of crossing fibres), interpretations based on single fibre experiments may not be appropriate. Our results suggest that variations in body composition relate to changes in fornix microstructure. However, in order to make inferences about the biophysical basis of these changes more direct microstructural indices of, for instance, axonal diameter  or density  will have to be employed in future studies.
The present study reports a correlational result and thus cannot infer direction of potential causality. Neurodevelopmental and inherited factors might predispose an individual to less well developed fornix connections and, hence, to less efficient communication and control within food regulatory circuits, leading to increased food intake. On the other hand, it may also be possible that body fat itself has detrimental effects on fornix white matter leading to increased vulnerability to pathological mechanisms associated, for instance, much later with exaggerated dementia risk . If this account is correct, then these effects must be anatomically specific given the lack of correlations in the parahippocampal cingulum.
The question of directionality as well as the functional significance of our result and the role of other potential factors such as hormonal modulation (e.g. cortisol)  need to be addressed in future prospective studies. In the present context, we might expect BMI associated changes in fornix AD to relate to the neurophysiological and psychological correlates of food intake control such as differential responding to food cues in satiety or food restricted states that could be studied with functional neuroimaging.
In summary, this is the first study to demonstrate a robust relationship between microstructural white matter indices in the fornix tract and variation in BMI within normal and overweight range in a group of older adults. These results have important implications for the understanding of neural factors contributing to an individual’s susceptibility to weight gain.
Conceived and designed the experiments: CMB MJO JPA DKJ. Performed the experiments: CMB. Analyzed the data: CMB RJB. Wrote the paper: CMB MJB RJB JPA DKJ.
- 1. Saunders RC, Aggleton JP (2007) Origin and topography of fibers contributing to the fornix in macaque monkeys. Hippocampus 17: 396–411. doi: 10.1002/hipo.20276
- 2. Burgess N, Maguire EA, O’Keefe J (2002) The human hippocampus and spatial and episodic memory. Neuron 35: 625–41. doi: 10.1016/s0896-6273(02)00830-9
- 3. Chudasama Y, Doobay VN, Liu Y (2012) Hippocampal-prefrontal cortical circuit mediates inhibitory response control in the rat. The Journal of Neuroscience 32: 10915–10924. doi: 10.1523/jneurosci.1463-12.2012
- 4. Davidson TL, Kanoski SE, Schier LA, Clegg DJ, Benoit SC (2007) A potential role for the hippocampus in energy intake and body weight regulation. Curr Opin Pharmacol 7: 613–6. doi: 10.1016/j.coph.2007.10.008
- 5. Davidson TL, Chan K, Jarrard LE, Kanoski SE, Clegg DJ, et al. (2009) Contributions of the hippocampus and medial prefrontal cortex to energy and body weight regulation. Hippocampus 19: 235–52. doi: 10.1002/hipo.20499
- 6. Raji CA, Ho AJ, Parikshak NN, Becker JT, Lopez OL, et al. (2010) Brain structure and obesity. Hum Brain Mapp 31: 353–64. doi: 10.1002/hbm.20870
- 7. DelParigi A, Chen K, Salbe AD, Hill JO, Wing RR, Reiman EM, et al. (2004) Persistence of abnormal neural responses to a meal in postobese individuals. Int J Obes Relat Metab Disord 28: 370–7. doi: 10.1038/sj.ijo.0802558
- 8. Kim DH, Woods C, Seeley RJ (2012) Hypothalamic Akt/PKB signalling in regulation of food intake. Frontiers in Bioscience 4: 953–66. doi: 10.2741/s311
- 9. Poletti CE, Cresswell G (1977) Fornix system efferent projections in the squirrel monkey: An experimental degeneration study. J Comp Neurol 175: 101–128. doi: 10.1002/cne.901750107
- 10. Blatt GJ, Rosene DL (1998) Organization of direct hippocampal efferent projections to the cerebral cortex of the rhesus monkey: Projections from CA1, prosubiculum, and subiculum to the temporal lobe. J Comp Neurol 393: 92–114. doi: 10.1002/(sici)1096-9861(19980302)392:1<92::aid-cne7>3.0.co;2-k
- 11. Friedman DP, Aggleton JP, Saunders RC (2002) A comparison of hippocampal, amygdala, and perirhinal projections to the nucleus accumbens: a combined anterograde and retrograde tracing study in the macaque brain. J Comp Neurol 450: 345–365. doi: 10.1002/cne.10336
- 12. Haber SN, Knutson B (2010) The Reward Circuit: Linking Primate Anatomy and Human Imaging Neuropsychopharmacology Reviews. 35: 4–26. doi: 10.1038/npp.2009.129
- 13. Schoenbaum G, Esber GR (2010) How do you (estimate you will) like them apples? Integration as a defining trait of orbitofrontal function. Curr Opin Neurobiol 20: 205–11. doi: 10.1016/j.conb.2010.01.009
- 14. Basser PJ, Pajevic S, Pierpaoli C, Duda J, Aldroubi A (2000) In vivo fiber tractography using DT-MRI data. Magn Reson Med 44: 625–32. doi: 10.1002/1522-2594(200010)44:4<625::aid-mrm17>3.0.co;2-o
- 15. Catani M, Howard RJ, Pajevic S, Jones DK (2002) Virtual in vivo interactive dissection of white matter fasciculi in the human brain. NeuroImage 17: 77–94. doi: 10.1006/nimg.2002.1136
- 16. Cabeza R, Ciaramelli E, Olson IR, Moscovitch M (2008) The parietal cortex and episodic memory: an attentional account. Nat Rev Neurosci 9: 613–625. doi: 10.1038/nrn2459
- 17. Jones DK, Christiansen K, Chapman RJ, Aggleton JP (2013) Distinct subdivisions of the cingulum bundle revealed by diffusion MRI fibre tracking: Implications for neuropsychological investigations. Neuropsychologia 51: 67–78. doi: 10.1016/j.neuropsychologia.2012.11.018
- 18. Jones DK, Cercignani M (2010) Twenty-five pitfalls in the analysis of diffusion MRI data. NMR Biomed 23: 803–820. doi: 10.1002/nbm.1543
- 19. Ikram MA, Vrooman HA, Vernooij MW, van der Lijn F, Hofman A, et al. (2008) Brain tissue volumes in the general elderly population. The Rotterdam Scan Study. Neurobiol Aging 29: 882–890. doi: 10.1016/j.neurobiolaging.2006.12.012
- 20. Metzler-Baddeley C, O’Sullivan MJ, Bells S, Pasternak O, Jones DK (2012) How and how not to correct for CSF-contamination in diffusion MRI. Neuroimage 59: 1394–403. doi: 10.1016/j.neuroimage.2011.08.043
- 21. Pasternak O, Sochen N, Gur Y, Intrator N, Assaf Y (2009) Free water elimination and mapping from diffusion MRI. Magn Reson Med 62: 717–30. doi: 10.1002/mrm.22055
- 22. Fazekas F, Kleinert R, Offenbacher H, Schmidt R, Kleinert G, et al. (1993) Pathological correlates of incidental MRI white matter signal hyperintensities. Neurology 43: 1683–1689. doi: 10.1212/wnl.43.9.1683
- 23. Nelson HE (1991) The National Adult Reading Test-Revised (NART-R): test manual. Windsor, UK: National Foundation for Educational Research–Nelson.
- 24. Metzler-Baddeley C, Jones DK, Belaroussi B, Aggleton JP, O’Sullivan MJ (2011) Frontotemporal connections in episodic memory and aging: a diffusion MRI tractography study. J Neurosci 31: 13236–45. doi: 10.1523/jneurosci.2317-11.2011
- 25. Sheikh JI, Yesavage JA (1986) Geriatric Depression Scale (GDS): recent evidence and development of a shorter version. In: Clinical gerontology: a guide to assessment and intervention, 165–173. New York: Haworth.
- 26. Jones DK, Simmons A, Williams SC, Horsfield MA (1999) Non-invasive assessment of axonal fiber connectivity in the human brain via diffusion tensor MRI. Magn Reson Med 42: 37–41. doi: 10.1002/(sici)1522-2594(199907)42:1<37::aid-mrm7>3.0.co;2-o
- 27. Leemans A, Jeurissen B, Siibers S, Jones DK (2009) ExploreDTI: A graphical tool box for processing, analyzing, and visualizing diffusion MR data. In 17th Anuual Meeting of the International Society of Magnetic Resonance in Medicine. Hawaii, USA.
- 28. Leemans A, Jones DK (2009) The B-matrix must be rotated when correcting for subject motion in DTI data. Magn Reson Med 61: 1336–49. doi: 10.1002/mrm.21890
- 29. Tournier JD, Calamante F, Gadian DG, Connelly A (2004) Direct estimation of the fiber orientation density function from diffusion-weighted MRI data using spherical deconvolution. Neuroimage 23: 1176–85. doi: 10.1016/j.neuroimage.2004.07.037
- 30. Jeurissen B, Leemans A, Jones DK, Tournier JD, Sijbers J (2011) Probabilistic tractography using the residual bootstrap with constrained super-resolved spherical harmonic deconvolution. Human Brain Mapping 32: 461–79. doi: 10.1002/hbm.21032
- 31. Pajevic S, Pierpaoli C (1999) Color schemes to represent orientation of anisotropic tissues from diffusion tensor data: Application to white matter fiber tract mapping in the human brain. Magn Reson Med 42: 526–540. doi: 10.1002/(sici)1522-2594(199909)42:3<526::aid-mrm15>3.3.co;2-a
- 32. Wakana S, Jiang H, Nagae-Poetscher LM, van Zijl PC, Mori S (2004) Fiber tract-based atlas of human white matter anatomy. Radiology 230: 77–87. doi: 10.1148/radiol.2301021640
- 33. Belaroussi B, O’Sullivan M, Vincent F, Pachai C (2008) HippoQuant: combining geometrical and intensity information for 3D hippocampus detection in 3D T1-weighted MRI images. Paper presented at 16th Scientific Meeting and Exhibition of the International Society in Magnetic Resonance in Medicine, Toronto, May.
- 34. Pruessner JC, Li LM, Serles W, Pruessner M, Collins DL, et al. (2000) Volumetry of hippocampus and amygdala with high resolution MRI and three-dimensional analysis software: minimising the discrepancies between laboratories. Cereb Cortex 10: 433–442. doi: 10.1093/cercor/10.4.433
- 35. Ashburner J, Friston KJ (2005) Unified segmentation. Neuroimage 26: 839–851. doi: 10.1016/j.neuroimage.2005.02.018
- 36. Grober E, Buschke H (1987) Genuine memory deficits in dementia. Dev Neuropsychol 3: 13–36. doi: 10.1080/87565648709540361
- 37. Sergi G, De Rui M, Coin A, Inelmen EM, Manzato E (2012) Weight loss and Alzheimer’s disease: temporal and aetiologic connections. Proc Nutr Soc 31: 1–6. doi: 10.1017/s0029665112002753
- 38. IBM SPSS Inc. (2011) PASW STATISTICS 18.0 Command Syntax Reference. IBM SPSS Inc., Chicago.
- 39. Benoit SC, Davis JF, Davidson TL (2010) Learned and cognitive controls of food intake. Brain Res. 1350: 71–6. doi: 10.1016/j.brainres.2010.06.009
- 40. Kringelbach ML (2005) The human orbitofrontal cortex: linking reward to hedonic experience. Nature Reviews Neuroscience 6: 691–702. doi: 10.1038/nrn1747
- 41. Roesch MR, Bryden DW, Cerri DH, Haney ZR, Schoenbaum G (2012) Willingness to Wait and Altered Encoding of Time-Discounted Reward in the Orbitofrontal Cortex with Normal Aging. The Journal of Neuroscience 32: 5525–5533. doi: 10.1523/jneurosci.0586-12.2012
- 42. Stanek KM, Grieve SM, Brickman AM, Korgaonkar MS, Paul RH, et al. (2011) Obesity is associated with reduced white matter integrity in otherwise healthy adults. Obesity 19: 500–4. doi: 10.1038/oby.2010.312
- 43. Assaf Y, Pasternak O (2008) Diffusion tensor imaging (DTI)-based white matter mapping in brain research: a review. Journal of Molecular Neuroscience 34: 51–61. doi: 10.1007/s12031-007-0029-0
- 44. Beaulieu C (2002) The basis of anisotropic water diffusion in the nervous system - a technical review. NMR Biomed 15: 435–55. doi: 10.1002/nbm.782
- 45. Assaf Y, Blumenfeld-Katzir T, Yovel Y, Basser PJ (2008) AxCaliber: a method for measuring axon diameter distribution from diffusion MRI. Magn Reson Med 59: 1347–54. doi: 10.1002/mrm.21577
- 46. Assaf Y, Basser PJ (2005) Composite hindered and restricted model of diffusion (CHARMED) MR imaging of the human brain. Neuroimage 27: 48–58. doi: 10.1016/j.neuroimage.2005.03.042
- 47. Anstey KJ, Cherbuin N, Budge M, Young J (2011) Body mass index in midlife and late-life as a risk factor for dementia: a meta-analysis of prospective studies. Obes Rev 12: 426–37. doi: 10.1111/j.1467-789x.2010.00825.x
- 48. Brown ES, Varghese FP, McEwen BS (2004) Association of depression with medical illness: does cortisol play a role. Biological Psychiatry 55: 1–9. doi: 10.1016/s0006-3223(03)00473-6