Abstract
Neuronal iron dyshomeostasis occurs in multiple neurodegenerative diseases. Changes in the Fe(II)/Fe(III) ratio toward Fe(II) is closely related to oxidative stress, lipid peroxidation, and represents a hallmark feature of ferroptosis. In particular for body fluids, like cerebrospinal fluid (CSF), reliable quantitative methods for Fe(II)/(III) redox-speciation analysis are needed to better assess the risk of Fe(II)-mediated damage in brain tissue. Currently in the field of metallomics, the most direct method to analyze both iron species is via LC-ICP-MS. However, this Fe(II)/(III) speciation analysis method suffers from several limitations. Here, we describe a unique method using capillary electrophoresis (CE)-ICP-MS for quantitative Fe(II)/(III) speciation analysis that can be applied for cell lysates and biofluid samples. Compared to LC, CE offers various advantages: (1) Capillaries have no stationary phase and do not depend on batch identity of stationary phases; (2) Replacement of aged or blocked capillaries is quick with no performance change; (3) Purge steps are effective and short; (4) Short sample analysis time. The final method employed 20 mM HCl as background electrolyte and a separation voltage of +25 kV. In contrary to the LC-method, no complexation of Fe-species with pyridine dicarboxylic acid (PDCA) was applied, since it hampered separation. Peak shapes and concentration detection limits were improved by combined conductivity-pH-stacking achieving 3 μg/L detection limit (3σ) at 13 nL injection volume. Calibrations from LOD-150 μg/L were linear [r 2 [Fe(II)] = 0.9999, r 2 [Fe(III)] = 0.9951]. At higher concentrations Fe(II) curve flattened significantly. Measurement precision was 3.5% [Fe(II) at 62 μg/L] or 2.2% [Fe(III) at 112 μg/L] and migration time precision was 2% for Fe(III) and 3% for Fe(II), each determined in 1:2 diluted lysates of human neuroblastoma cells. Concentration determination accuracy was checked by parallel measurements of SH-SY5Y cell lysates with validated LC-ICP-MS method and by recovery experiments after standard addition. Accuracy (n = 6) was 97.6 ± 3.7% Fe(III) and 105 ± 6.6%Fe(II). Recovery [(a) +33 μg/L or (b) +500 μg/L, addition per species] was (a): 97.2 ± 13% [Fe(II)], 108 ± 15% [Fe(III)], 102.5 ± 7% (sum of species), and (b) 99±4% [Fe(II)], 101 ± 6% [Fe(III)], 100 ± 5% (sum of species). Migration time shifts in CSF samples were due to high salinity, but both Fe-species were identified by standard addition.