Oxidized Cell-Free Hemoglobin Induces Mitochondrial Dysfunction by Activation of the Mitochondrial Permeability Transition Pore in the Pulmonary Microvasculature

Abstract

Objective: Cell-free hemoglobin (CFH) is released into the circulation during sepsis where it can redox cycle from the ferrous 2+ to ferric 3+ and disrupt endothelial function, but the mechanisms of CF-mediated endothelial dysfunction are unknown. We hypothesized that oxidized CFH induces mitochondrial dysfunction via the mitochondrial permeability transition pore (mPTP) in pulmonary endothelial cells, leading to the release of mitochondrial DNA (mtDNA).

Methods: Human lung microvascular endothelial cells were treated with CFH2+/CFH3+. We measured mitochondrial mPTP activation (flow cytometry), network and mass (immunostaining), structure (electron microscopy), mtDNA release (PCR), and oxygen consumption rate (OCR; Seahorse). Plasma from critically ill patients and conditioned cell media were quantified for mtDNA and CFH.

Results: CFH3+ disrupted the mitochondrial network, activated the mPTP (1434 (874-1642) vs. 2302 (1729-2654) mean fluorescent intensity, p = 0.02), increased the spare respiratory capacity (30.61 (29.36-37.78) vs. 7.83 (3.715-10.63) OCR, p = 0.004), and caused the release of mtDNA. CFH was associated with circulating mtDNA (R² = 0.1912, p = 0.0077) in plasma from critically ill patients.

Conclusion: CFH3+, not CFH2+, is the primary driver of CFH-induced lung microvascular mitochondrial dysfunction. Activation of the mPTP and the release of mtDNA are a feature of CFH3+ mediated injury.

Keywords: cell‐free hemoglobin; mitochondrial dysfunction; mitochondrial permeability transition pore; mtDNA.

FIGURE 1

CFH3+ induces mitochondrial superoxide and cellular ROS formation. MitoSOX stained HLMVECs were treated for the indicated time points before being analyzed on a BD LSRFortessa in the PE channel. (A) Quantification by background adjusted MFI of MitoSOX staining of HLMVECs at the time points of 1, 6, and 24 h. CellROX stained HLMVECs were imaged on a BioTek Lionheart FX microscope using the Cy5 filter set (628/685). (B) Quantification of CellROX Deep Red staining at the time points of 1, 6, and 24 h. (C) Pre‐treatment with Hp for 1 h reduced CellROX staining from CFH3+ exposure at 24 h (Mann–Whitney test). (A,C) n = 6 from 3 donors (two male, one female), (B) n = 7–15 from 4 donors (two male, two female). * = p < 0.05, ** = p < 0.01 , and **** = p < 0.0001. Graphs show median with interquartile range.

FIGURE 2

Treatment of endothelial monolayers with CFH3+ increases maximal respiration and spare respiratory capacity. Confluent HLMVEC monolayers were treated for 6 h with CFH before room temp equilibration prior to analysis on the Seahorse Analyzer using the Mito Stress Test. Hp pretreatment was for 1 h for the VEH + Hp1‐1 and CFH3++Hp1‐1 groups. (A) Normalized OCR data from donor 57F showing the change in oxygen consumption after the addition of oligomycin (~20 min), FCCP (~40 min), and rotenone and antimycin A (~60 min). Quantification of metabolic parameters of (B) basal respiration, (C) maximal respiration, (D) spare respiratory capacity, (E) proton leak, and (F) ATP production. n = 8 from 5 donors (three male, two female) with some donors repeated. ** = p < 0.01.

FIGURE 3

CFH3+ induces mitochondrial morphological changes. (A) Quantification of the mitochondrial footprint per field of view using the Mitochondrial Network Analysis (MiNA) ImageJ plugin. Representative images of the MiNA output for (B) vehicle, (C) CFH2+, (D) CFH3+, and (E) Paraquat treated cells with nuclei highlighted (white arrows; scale bar = 10 um). Representative TEM micrographs for (F) vehicle, (G) CFH2+, and (H) CFH3+ treated HLMVECs with circularity demonstration (red circles) and mitochondria highlighted (black arrows; scale bar = 400 nm). Quantification of mitochondrial morphology parameters of (I) mean electron density, (J) circularity (4pi (area/perimeter²)) and (K) aspect ratio (major axis: Minor axis). (A–E) n = 5, (F, vehicle) n = 228, (G, CFH2+) n = 163, (H, CFH3+) n = 416. * = p < 0.05, ** = p < 0.01 , and **** = p < 0.0001.

FIGURE 4

CFH3+ decreases mitochondrial mass. (A) Quantification of MitoGreen by object MFI based on the DNA counterstain. Representative images (Blue = nuclei, Green = mitochondria; scale bar = 2000 um) from HLMVECs stained with MitoTracker Green and treated with (B) vehicle and CFH3+ at the (C) 1, (D) 3, and (E) 6 h time points. n = 6 from 3 donors (two male, one female). * = p < 0.05, *** = p < 0.001, **** = p < 0.0001.

FIGURE 5

CFH causes partial depolarization of mitochondria. (A) Quantification of the shift in fluorescence by the ratio of PE:FITC channels. Representative stacked histograms of the JC10 stained cells in the (B) PE and (C) FITC filter sets. n = 12 from 3 donors (two male, one female). ** = p < 0.01 and *** = p < 0.001.

FIGURE 6

CFH3+ activates the mPTP and induces release of mtDNA. (A) Quantification of HLMVECs stained with calcein‐AM and cobalt chloride measured by MFI. (B) Representative stacked histogram of the FITC fluorescence shift from CFH3+ treatment. (C) Quantification of the change in mtDNA release in the media supernatant by qPCR. (D) Quantification of the circulating mtDNA copy number by ddPCR in patient plasma in association with CFH levels. Probes are specific to the mt‐ND4L gene. (A) n = 7 from 3 donors (two male, one female), (C) n = 14 from 5 donors (three male, two female), and (D) n = 36. * = p < 0.05.