Laser ablation-inductively coupled plasma-mass spectrometry imaging of white and gray matter iron distribution in Alzheimer's disease frontal cortex

Highlights

• Imaging of phosphorus and iron by LA-ICP-MS defines white and gray matter iron.

• Iron is elevated in Alzheimer’s disease frontal cortex gray matter.

• LA-ICP-MS imaging reveals subcortical banding of iron.

• Iron levels appear to be slightly increased in white matter of Alzheimer’s patients.

Abstract

Iron deposition in the brain is a feature of normal aging, though in several neurodegenerative disorders, including Alzheimer's disease, the rate of iron accumulation is more advanced than in age-matched controls. Using laser ablation-inductively coupled plasma-mass spectrometry imaging we present here a pilot study that quantitatively assessed the iron content of white and gray matter in paraffin-embedded sections from the frontal cortex of Alzheimer's and control subjects. Using the phosphorus image as a confirmed proxy for the white/gray matter boundary, we found that increased intrusion of iron into gray matter occurs in the Alzheimer's brain compared to controls, which may be indicative of either a loss of iron homeostasis in this vulnerable brain region, or provide evidence of increased inflammatory processes as a response to chronic neurodegeneration. We also observed a trend of increasing iron within the white matter of the frontal cortex, potentially indicative of disrupted iron metabolism preceding loss of myelin integrity. Considering the known potential toxicity of excessive iron in the brain, our results provide supporting evidence for the continuous development of novel magnetic resonance imaging approaches for assessing white and gray matter iron accumulation in Alzheimer's disease.

Introduction

Disrupted iron metabolism appears to be a pathological hallmark in the Alzheimer's disease brain (Roberts et al., 2011). Numerous studies have identified abnormal increases in the iron concentration within a range of affected brain regions (Belaidi and Bush, 2015), including accumulation on the β-amyloid senile plaques that are characteristic of AD (Lovell et al., 1998). While a 2011 meta-analysis suggested a possible citation bias has overstated the significance of iron elevation (Schrag et al., 2011), it should not be ignored that disrupted iron homeostasis without a measurable increase still has the potential to promote oxidative stress through improper redox-silencing of this highly reactive species. Changes in chemical properties of brain iron have been observed dating back over half a century (Hallgren and Sourander, 1960), and contemporary biotechnology has identified a range of genetic and metabolic factors that support iron dyshomeostasis as playing an important role in AD pathology (Crespo et al., 2014).

Important iron regulatory proteins, including ferritin and transferrin appear to be both dysfunctional and abnormally distributed in the AD brain (Connor, J.R., et al., 1992a, Connor, J.R., et al., 1992b, Connor, J.R., et al., 1995), potentially contributing to the reactive ‘labile iron pool’ through mismanagement of normal metabolic pathways. Neuroinflammation, where glial cells promote the deposition of iron, contributes to elevated oxidative stress and mitochondrial dysfunction, and may also promote the aggregation of the β-amyloid peptide and tau protein, forming the plaques and tangles characteristic of the disease (Ong and Farooqui, 2005). Combined with the natural accumulation of iron in the aging brain, endogenous response to elevated cortical iron (such as heme oxygenase-1, which degrades heme and can release free, reactive ferrous [Fe^2 +] iron) may represent an important biochemical mechanism preceding neuronal damage in AD (Ward et al., 2014).

In vivo imaging of the AD brain using magnetic resonance imaging (MRI) has provided useful insight into both structural changes (Bartzokis et al., 2003) and iron deposition (Bartzokis, G., et al., 2000, Langkammer, C., et al., 2014), using techniques such as R₂ and R₂* relaxometry (Langkammer et al., 2010) and phase imaging (Zhu et al., 2009). However, differentiation between white and gray matter iron distribution in the neocortex using MRI is challenging, as typical MRI approaches are not absolutely quantitative; there are multiple contributions to tissue contrast (including myelin, iron and CSF); and most MRI methods have a spatial resolution that precludes fine detail definition of brain iron distribution at micrometer scales. Because of these many limitations, MR imaging of brain iron has been largely constrained to deep brain nuclei, such as the basal ganglia, which contain the highest iron content throughout the brain.

In this study we employed quantitative iron imaging by laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS) to compare the distribution of iron in white and gray matter regions of postmortem AD and healthy control (HC) frontal cortex tissue which are primarily affected by AD pathology. LA-ICP-MS employs a focused beam (typically in the ultra-violet range) that ablates particles from the tissue sample surface, which are then carried to the ICP-MS and measured on the basis of mass-to-charge (m/z) ratio (Hare et al., 2015). LA-ICP-MS is highly specific and sensitive to iron, with detection limits well below the typical biological concentrations found in neurological tissue (O'Reilly et al., 2014). With appropriate signal normalization and periodic sampling of standards with comparable matrix composition, LA-ICP-MS can provide absolute quantitative information at the low micrometer scale (1–100 + μm) (Hare, D.J., et al., 2012a, Miliszkiewicz, N., et al., 2015). As an element-specific detector, LA-ICP-MS also permits simultaneous detection of multiple analytes and generation of hyperspectral images. We exploited this capability here by using phosphorus distribution as a proxy for white and gray matter, which was then applied to differentiating iron distribution in the two regions of frontal cortex tissue from both AD and HC brains.