It has been proposed that increased tissue iron concentration, which has been observed in certain regions of the brain in individuals with Alzheimer’s Disease (AD), could provide a marker for diagnosis through detection with MRI. This is investigated in this thesis using high field MRI to examine post mortem human brain tissue. It is shown here that by using data from multiple brain regions discriminant analysis can successfully differentiate between AD and control samples, even when no statistically significant differences are observed in individual brain regions. A unique set of complementary techniques was used to investigate iron content, R2 and R2* of tissue samples from the caudate nucleus, putamen, globus pallidus substantia nigra, amygdala and pons, from a set of three control and AD cases. The particulate iron content of the samples was investigated by SQUID magnetometry and was followed by iron quantification. A trend of increased particulate and total iron concentration was observed in the AD tissue compared to control, however this did not reach statistical significance in any brain region. High resolution MRI relaxometry at 9.4 T was carried out on tissue from the caudate nucleus, putamen, globus pallidus and substantia nigra using a custom design Bruker micro-imaging MicWB40 probe. As part of the work towards this PhD the probe was tested, and MRI relaxometry protocols optimised for high resolution (86 x 86 μm in plane, 150 μm slice thickness) mapping of tissue samples with high iron concentration. Again, no statistically significant differences were observed between AD and control tissue. However, discriminant analysis of these data (particulate or total iron or R2 or R2*) from multiple brain regions achieved differentiation between control and AD cases with 100% sensitivity and specificity for this small sample set. This demonstrates the potential clinical usefulness of MRI of measurements of non-haem brain iron to aid in disease diagnosis. Synchrotron X-ray fluorescence (SXRF) mapping of 30 μm thick sections, cut from the MRI samples, showed the relative concentration distribution of iron, copper and zinc in one AD and control sample from each brain region. Each metal was shown to have a distinct distribution. In particular, the inhomogeneity of iron concentration within individual brain regions, such as the putamen, was demonstrated. This may explain the wide variation in iron concentration reported in the literature for the same brain regions, and highlights the importance of close anatomical matching of samples when making comparisons. The ability of high resolution SXRF mapping to investigate the metal content within individual cells was demonstrated and used to show an increase in iron in individual AD neurons, in addition to the surrounding grey and white matter tissue. Spatially matched SXRF and MRI maps were used to demonstrate a strong, statistically significant linear relationship between tissue iron concentration and R2, R2* and R2’ at 9.4 T. The gradient of the linear relationship between iron and R2, agrees extremely well with the predicted gradient at this field, where the prediction was made by Vymazal et al. (1996) using MRI relaxometry at 0.05 to 1.5 T. To the best of my knowledge, this is the first time that this relationship has been quantified at 9.4 T, or at any field above 7 T. MRI at 14.1 T was carried out on low iron concentration regions (the pons and amygdala). Matched SXRF and R2* maps did not show a strong linear relationship between iron and R2*. The iron concentration in these regions is less than 50 μg/g and it was concluded that in tissue with low iron content, other tissue properties - such as water content - are dominating the value of R2*. This result was replicated with data measured at 9.4 T, when only tissue with an iron concentration of less than 50 μg/g was considered.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:606115 |
| Date | January 2013 |
| Creators | Finnegan, Mary E. |
| Publisher | University of Warwick |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://wrap.warwick.ac.uk/60293/ |
Page generated in 0.0022 seconds