4596

Evidence of a link between brain structure and function and gut permeability: a combined RS-fMRI and DTI investigation of the brain-gut axis in healthy women
Suzanne T Witt1, Olga Bednarska2, Adriane Icenhour1,2, Sigrid Elsenbruch3, Johan D Söderholm2, Maria Engström1,4, Emeran A Mayer5, Åsa Keita2, and Susanna Walter1,2

1CMIV, Linöping University, Linköping, Sweden, 2Dept. of Clinical and Experimental Medicine, Linköping University, 3Institute of Medical Psychology & Behavioral Immunobiology, University of Duisburg-Essen, Essen, Germany, 4Dept. of Medical and Health Sciences, Linköping University, 5Dept. of Medicine, UCLA, Los Angeles, CA, United States

Synopsis

The brain-gut axis is thought to play a key role in the regulation of the gastrointestinal system with overall physical and emotional health. In diseases, such as irritable bowel syndrome, dysfunction within brain-gut interactions have been proposed to underlie symptoms of chronic abdominal pain. This study demonstrated that variations in gut mucosal permeability affected both resting-state functional connectivity in the DMN and white matter microstructure properties in healthy adult women. Variations within brain function and structure were apparent even when variations in gut permeability were small and remained within the normal range, indicating that brain-gut interactions may be quite sensitive.

INTRODUCTION

The brain-gut axis is thought to play a key role in the regulation of the gastrointestinal system with overall physical and emotional health. In diseases, such as irritable bowel syndrome, dysfunction within brain-gut interactions have been proposed to underlie symptoms of chronic abdominal pain.1 Changes in gut mucosal permeability have been implicated in sensitizing pain pathways.2 Similarly, changes in resting-state functional connectivity, especially in the default mode network, have been shown to change in patients reporting chronic pain.3 To date, though, no study has investigated brain-gut interactions in either healthy or diseased populations using organic measures of both gut and brain structure and function. The aim of the current study was to examine the relationships between in vitro measures of gut mucosal permeability and resting state functional connectivity in the default mode network and whole brain white matter microstructure in a population of healthy women. We anticipated observing that increases in gut mucosal permeability would be related to increased functional and structural connectivity among brain regions responsible for processing pain signals in the brain.

METHODS

Fifteen healthy women were included in the study (mean age = 29.7 yrs). Ten minutes of eyes closed resting state fMRI data (TR/TE = 2000/37ms; voxel = 3.59x3.59x4mm3; 28 slices; 300 time points) and diffusion weighted MRI data (TR/TE = 9339/82ms; voxel = 2mm3; 64 directions; b = 1000) were acquired on a 3T scanner (Philips Ingenia). FMRI data were preprocessed (realignment, normalization to MNI space, smoothing with 8mm FWHM Gaussian kernel) in SPM8. The GIFT group independent component analysis toolbox was used to identify the default mode network (DMN). The diffusion data were processed using standard procedures in FSL, and group-level maps of fractional anisotropy (FA) were calculated using tract based spatial statistics (TBSS). Colonic biopsies were acquired within two weeks post-MRI scanning after 8 hours of fasting. Biopsy samples were mounted in Ussing chambers4,5, and a paracellular probe (51chromium (Cr)-EDTA) was added to the mucosal side of the chambers. Passage of 51Cr-EDTA through the biopsies to the serosal side of the chambers was measured by gamma counting after 60 and 120 minutes (Figure 1). Whole brain covariate analyses were performed in PALM6 for the DMN and FA data separately to determine the relationship between paracellular passage of 51Cr-EDTA and DMN functional connectivity and FA values, respectively.

RESULTS

Variations in gut mucosal paracellular permeability correlated with significant whole-brain variations in resting state functional connectivity in the DMN (Figure 2) and white matter microstructure as measured by FA (Figure 3). Increases in paracellular permeability correlated with increased connectivity between the DMN and a number of regions thought to comprise the ascending and descending pain pathways in the brain, including thalamus, supplementary motor area, right middle frontal gyrus, raphe nucleus magnus, and bilateral post-central gyri. Decreases in paracellular permeability also correlated with increased connectivity between DMN and pain processing related brain regions, including ACC, right insula, bilateral precentral gyri, and periaqueductal grey. For white matter microstructure, increases in paracellular permeability in the gut correlated with increased FA in a number of major tracts, including corticospinal, corticothalamic, anterior thalamic radiations, corpus callosum, superior longitudinal fasciculi, as well as a number of association fibers. Decreases in paracellular permeability only appeared to correlate predominantly with increased FA in association fibers.

CONCLUSIONS

This study has demonstrated that variations in gut mucosal permeability may affect both resting-state functional connectivity in the DMN and white matter microstructure properties in healthy adult women. In the case of increases in paracellular permeability, the results indicate good anatomical correspondence between the white matter tracts exhibiting increased FA and the brain regions exhibiting increased functional connectivity, suggesting a structure-function match. The results also indicate that gut mucosal paracellular permeability is processed as pain. The results further suggest that variations within the brain are apparent even when variations in gut permeability are small and remain within the range for gut mucosal paracellular permeability previously reported in healthy adults5, indicating that brain-gut interactions may be quite sensitive. It is, however, still unclear whether these variations are temporary and/or modifiable, or whether variations in gut permeability precede variations in brain function and structure or vice versa.

Acknowledgements

No acknowledgement found.

References

1Mayer EA and Tillisch K, Annual Review of Medicine, 2011; 6:381-96.

2Vergnolle N et al., Nature Medicine, 2001; 7: 821-6.

3Farmer MA et al., Neuroscience Letters, 2012; 520: 197-203.

4Grass GM et al., Pharmacological Research, 1988; 5: 372-6.

5Keita AV et al., Laboratory investigation: a journal of technical methods and pathology, 2010; 59: 1213-21.

6Winkler AM et al., NeuroImage, 2014; 92: 381-97.

Figures

Figure 1. Paracellular permeability, as measured by 51Cr-EDTA passage, for the fifteen healthy volunteers. Passage given in units of cm/s x 10-6.

Figure 2. Whole-brain covariate analysis results showing brain regions whose connection to the default mode network depends on paracellular permeability. Brain regions with increased connectivity with the default mode network with increasing paracellular permeability shown in red-yellow. Brain regions with increased connectivity with the default mode network with decreasing paracellular permeability shown in blue-green. Results scaled in terms of –log10(p), thresholded at p < 0.05, Family Wise Error correction for comparing across the whole brain.

Figure 3. Whole-brain covariate analysis results showing white matter tracts whose microstructural properties depend on paracellular permeability. Tracts with increased fractional anisotropy with increasing paracellular permeability shown in red-yellow. Tracts with increased fractional anisotropy with decreasing paracellular permeability shown in blue-green. Results scaled in terms of –log10(p), thresholded at p < 0.05, Family Wise Error correction for comparing across the whole brain.

Proc. Intl. Soc. Mag. Reson. Med. 25 (2017)
4596