Lipid peroxidation in diamond supported bilayers

Abstract

Lipid peroxidation is a process that occurs in cells when they are exposed to oxidative stress. During the process reactive oxygen species attack lipids within the lipid bilayers of cells. Since the products of lipid peroxidation are toxic and carcinogenic, it is important to understand where and how it occurs with nanoscale resolution. The radical intermediates of this process are particularly interesting since they are causing chain reactions damaging large parts of the lipid membranes in cells. However, they are also difficult to measure for the state of the art because they are short lived and reactive. Here, we study the lipid peroxidation of three artificial lipid bilayers on a diamonds substrate that can be used to study lipid peroxidation. In particular, we present a diamond quantum sensing method called T₁-relaxometry that allows for in situ measurements and imaging of radical intermediates of lipid peroxidation in these membranes.

Fig. 1. Diamond quantum relaxometry. (a) Physical structure of the nitrogen-vacancy (NV) center in diamond (b) T₁ relaxometry curve with and without the presence of paramagnetic species (in this case radical intermediates from lipid peroxidation).

Fig. 2. Overview of the planar lipid bilayer preparation and oxidation experiment. (a) Lipid structure and diamond sample with NV centres near the surface (b) Chemical incubation step. The lipid bilayer is grown right on top of the diamond sensor (c) Final structure. The lipid bilayer is supported in the diamond sensor (d) An oxidizing agent is added in order to initiate the lipid peroxidation process.

Fig. 3. Phospholipids used to detect lipid peroxidation in artificial membranes. (a) POPC (b) POPG and (c) DPPC. The lipids on the top have a double bond in the alkyl chain that allows for lipid peroxidation. (d) Mechanism of lipid peroxidation for POPG as an example.

Fig. 4. Experimental procedure. (a) Optical readout of the NV center and membrane system (simplified). (b) Pulsed sequence used in order to retrieve the T₁ relaxometry information. (c) Grouping of the images for the relaxometry imaging procedure, in which every pixel contains a relaxometry curve. (d) Grouping of the images for the lipid peroxidation detection. In this case, all the pixels are added together to retrieve a single relaxometry curve.

Fig. 5. Effect of the hydrogen peroxide in the longitudinal relaxation time as a function of the oxidizing agent concentration. Each experiment shows the T₁ relaxation time at different H₂O₂ concentrations, including without H₂O₂ (first experiment). Statistical significance (N = 6) was assessed with an unpaired t-test, showing statistical significant differences only at 15% concentration v/v.

Fig. 6. Relaxation times for the diamond and membrane systems, with (immediately after measuring control) and without oxidizing agent (control). Statistical significance (N = 6) was assessed with an unpaired t-test, having statistical significance only for the POPC and the POPG pairs.

Fig. 7. Relaxometry images of the different membranes (up) and the same membranes after exposure to H₂O₂ (bottom).

Fig. 8. Confocal imaging of the stained lipid bilayers. The bigger images have a size of 300 × 300 μm, while the zoom-in have a size of 150 × 150 μm.

Fig. 9. AFM images of the of the different membranes (up) and the same membranes after been exposed to H₂O₂ (bottom), alongside their roughness.

Fig. 10. Analysis of the roughness of the AFM images. Statistical significance (N = 5) was assessed with an unpaired t-test, having found statistical significance with all pairs. The bare diamond sample has a p < 0.001 (****) significant difference with all the pairs.

Fig. 11. XPS spectra for the membranes before and after H₂O₂. The row (a) on top shows the spectra for the samples before being exposed to the oxidizing agent, while the row (b) on bottom displays the spectra from the same sample after being exposed to H₂O₂.

Fig. 12. Atomic composition difference for all the membranes in the oxygen peak. Statistical significance (N = 10) was assessed with an unpaired t-test, having statistical significance only for the POPG pair.