Simultaneous Fe2+/Fe3+ imaging shows Fe3+ over Fe2+ enrichment in Alzheimer's disease mouse brain

Abstract

Visualizing redox-active metal ions, such as Fe²⁺ and Fe³⁺ ions, are essential for understanding their roles in biological processes and human diseases. Despite the development of imaging probes and techniques, imaging both Fe²⁺ and Fe³⁺ simultaneously in living cells with high selectivity and sensitivity has not been reported. Here, we selected and developed DNAzyme-based fluorescent turn-on sensors that are selective for either Fe²⁺ or Fe³⁺, revealing a decreased Fe³⁺/Fe²⁺ ratio during ferroptosis and an increased Fe³⁺/Fe²⁺ ratio in Alzheimer's disease mouse brain. The elevated Fe³⁺/Fe²⁺ ratio was mainly observed in amyloid plaque regions, suggesting a correlation between amyloid plaques and the accumulation of Fe³⁺ and/or conversion of Fe²⁺ to Fe³⁺. Our sensors can provide deep insights into the biological roles of labile iron redox cycling.

Fig. 1.. The sequences of the Fe²⁺- and Fe³⁺-specific DNAzymes and the design, sensitivity, and selectivity of their fluorescent sensors.

(A) Secondary structure of the trans-cleaving Fe²⁺-specific DNAzyme [Fe(II)-H5], consisting of an enzyme strand shown in green with an Iowa Black FQ quencher (Q1) at the 5′ end, and a substrate strand shown in black with the same quencher (Q2) at the 5′ end and an Alexa Fluor 488 fluorophore (F1) at the 3′ end. (B) Secondary structure of the trans-cleaving Fe³⁺-specific DNAzyme [Fe(III)-B12], consisting of an enzyme strand shown in purple with an Iowa Black RQ quencher (Q3) at the 3′ end, and a substrate strand shown in black with a 5′ Alexa Fluor 647 fluorophore (F2) and a second Iowa Black RQ quencher (Q4) at the 3′ end. Both DNAzymes contain a ribonucleotide cleavage site (red, arrow). In addition, a mutation in the enzyme strand shown in blue [A to G in the Fe(II)-H5 DNAzyme and T to C in the Fe(III)-B12 DNAzyme] rendered the respective DNAzymes inactive and thus served as negative controls. (C and D) Selectivity of the trans-cleaving Fe(II)-H5 (C) and Fe(III)-B12 (D) DNAzymes. Fraction of the cleaved substrate in the presence of different metal ions in 20 mM acetate buffer (pH 6.0), 5 mM bis-tris, and 200 mM NaCl analyzed via denaturing polyacrylamide gel electrophoresis (PAGE). (E and F) Normalized fluorescence intensity of the Fe(II)-H5 DNAzyme (E) and the Fe(III)-B12 DNAzyme (F) sensors in response to different concentrations of Fe²⁺ or Fe³⁺, respectively. (G and H) The initial fluorescence turn-on rates (k_obs) of the sensors at different iron concentrations are shown in (G) for the Fe(II)-H5 DNAzyme and in (H) for the Fe(III)-B12 DNAzyme.

Fig. 2.. Simultaneous imaging of labile Fe²⁺ and Fe³⁺ in HepG2 cells.

(A) HepG2 cells transfected with the Fe(II)-H5 (green) and the Fe(III)-B12 (red) DNAzyme sensors using PEI. Lysosomes are stained with LysoTracker Red and are shown in magenta. Cells were without any additional treatment to image endogenous (endo) iron, treated with 5 or 50 μM Transferrin (Tf) to increase lysosomal iron, or with 100 μM iron chelator DFO to decrease intracellular iron. (B) Statistical analysis of mean fluorescence intensity in LysoTracker-labeled region in (A) reveals that Fe²⁺ increased 1.9-fold when treated with 5 μM Tf and increased 4.6-fold when treated with 50 μM Tf, while Fe²⁺ decreased to barely detectable when treated with 100 μM DFO. (C) Statistical analysis of mean fluorescence intensity in (A) reveal that Fe³⁺ increased 1.6-fold when treated with 5 μM Tf and increased 1.9-fold when treated with 50 μM Tf and decreased around 5.7-fold when treated with 100 μM DFO. Cells were cultured in iron-deficient medium during the treatments. Error bars represent SEMs of the LysoTracker-labeled region in 20 different cells in each sample. Fluorescence intensity was normalized with ErS-iErS. Scale bar, 20 μm. (D) The Fe³⁺/Fe²⁺ ratio was calculated based on the mean fluorescence intensity of Fe²⁺ (B) and Fe³⁺ (C) in the Lysotracker region. When comparing with endogenous Fe³⁺/Fe²⁺ ratio in cells that were in the iron deficient media, no significant change in the Fe³⁺/Fe²⁺ ratio was observed in cells that were treated with different concentrations of Tf, and a 2.4-fold decrease of the Fe³⁺/Fe²⁺ ratio was observed in cells that were treated with 100 µM DFO.

Fig. 3.. Ferroptosis triggers an elevation in the intracellular Fe²⁺ and Fe³⁺ pools with a decreased Fe³⁺/Fe²⁺ ratio.

(A) Fe(II)-H5 (green) and Fe(III)-B12 (red) DNAzyme sensors detecting labile pools of Fe²⁺ and Fe³⁺ simultaneously during RSL3-induced ferroptosis at different time points. LysoTracker Red and Hoechst 33258 were used to label lysosomes (magenta) and nucleus (blue), respectively. (B and C) Statistical analysis of normalized fluorescence intensity in the LysoTracker-labeled region reveals an increase of Fe²⁺ (B) and Fe³⁺ (C) within the first 4 hours and a decrease of Fe²⁺ and Fe³⁺ within 4 to 8 hours of RSL3-induced ferroptosis in the endosomal-lysosomal system. (D) The Fe³⁺/Fe²⁺ ratio decreases during ferroptosis. Scale bar, 20 μm. h, hours.

Fig. 4.. Fe³⁺/Fe²⁺ ratio increases in brain regions containing amyloid plaques.

(A) Both Fe²⁺ and Fe³⁺ were monitored in 11-month-old mouse brain slices that immunostained with the HJ3.4 antibody, which labels immunoreactive Aβ plaques, to trace the colocalization between iron and Aβ plaques in mouse brains. ErS, the active version of iron sensors containing the enzyme (E) strand and the substrate strand with the ribonucleotide as the cleavage site (rS); iErS, the inactive DNAzyme sensor control, which contains a point mutation in the enzyme strand that renders the DNAzyme inactive in the presence of Fe²⁺ or Fe³⁺. The signal from iErS is considered as background signal that is caused by auto fluorescence or Fe²⁺- or Fe³⁺-independent cleavages (B and C) Statistical analysis of fluorescence intensity in (A). Fluorescent signals were normalized by subtracting the background detected with inactive DNAzyme sensors. Both Fe²⁺ (B) and Fe³⁺ (C) levels were elevated in 5xFAD mice when compared with wild-type (WT) controls. (D) When monitoring the distribution of Fe²⁺ and Fe³⁺ and comparing with WT mouse brains, Fe²⁺ level was elevated 2.1-fold in the cortex regions with Aβ plaques (APDR) and was elevated 1.7-fold in the surrounding cortex regions that did not show Aβ plaque deposition signal (non-APDR). (E) Fe³⁺ increased more (8.7-fold) in APDR than the non-APDR (2.6-fold). (F) The Fe³⁺/Fe²⁺ ratio showed increase in whole-brain slices of WT and 5xFAD mice. (G) Most of the increased Fe³⁺/Fe²⁺ ratio was in the APDR but not non-APDR in 5xFAD mouse cortex. Scale bar, 50 μm. Analyzed with paired Student’s t test. *P < 0.05, **P < 0.01, and ***P < 0.001. ns, not significant.