A fMRI study to decipher the regional effects of an intraperitoneal glucose dose in the fasted rat model

model (2 nd level analysis) was applied to test for changes in the BOLD signal between pre- and post-glucose images in the time-series. SPM{t}distribution was thresholded at p<0.001 (uncorrected for multiple comparisons). RESULTS AND DISCUSSIONS Plasma glucose concentrations was significantly higher after injection of glucose (223.2±53.4mg/dl, mean±sd) compared to saline (147.9±22.16mg/dl, P<0.05). Significant widespread increases in the BOLD-MRI signal after glucose administration were observed in the cerebellum, brainstem, other hindbrain and midbrain regions, hippocampus, hypothalamus, thalamus and striatum (Figure 1). In contrast, less extensive changes were observed following saline, with only small increases in the posterior cortex and (unilateral) hippocampus (data not shown). Both saline and glucose-treated animals showed significant BOLD-contrast decreases in the prefrontal cortex and olfactory regions, possibly related to i.p. volume administration. Significant activations were observed in the hypothalamic and hippocampal areas. The activations in the former are consistent with the known presence of glucosensing neurons in the ventromedial nucleus containing higher proportions of GE neurones 4 . Likewise, glucosensing neurones are also present in the hippocampus, the neurones becoming more depolarised, i.e. excitable with increasing glucose concentrations 5 . However, the changes observed following glucose dosing may also result from insulin action in the brain: insulinaemia, arising from glucose-stimulated release of insulin from pancreatic beta cells. We did not observe a transient decrease in activity in the hypothalamus previously reported 3,4 following i.p. glucose administration. Our methodology may be insensitive to this transient small decrease but sensitive to changes occurring throughout the brain. Further studies are needed to determine the source of the changes in the BOLD-MRI signal following i.p. glucose administration. CONCLUSION We have shown the ability to non-invasively assess changes in the brain following a glucose dose, suggesting the possibly use of this technique to determine the functional role of various nutrients under different physiological states.