An intra articular injectable Mitocelle recovers dysfunctional mitochondria in cellular organelle disorders

Abstract

Mitochondrial dysfunction increases ROS production and is closely related to many degenerative cellular organelle diseases. The NOX4-p22phox axis is a major contributor to ROS production and its dysregulation is expected to disrupt mitochondrial function. However, the field lacks a competitive inhibitor of the NOX4-p22phox interaction. Here, we created a povidone micelle-based Prussian blue nanozyme that we named "Mitocelle" to target the NOX4-p22phox axis, and characterized its impact on the major degenerative cellular organelle disease, osteoarthritis (OA). Mitocelle is composed of FDA-approved and biocompatible materials, has a regular spherical shape, and is approximately 88 nm in diameter. Mitocelle competitively inhibits the NOX4-p22phox interaction, and its uptake by chondrocytes can protect against mitochondrial malfunction. Upon intra-articular injection to an OA mouse model, Mitocelle shows long-term stability, effective uptake into the cartilage matrix, and the ability to attenuate joint degradation. Collectively, our findings suggest that Mitocelle, which functions as a competitive inhibitor of NOX4-p22phox, may be suitable for translational research as a therapeutic for OA and cellular organelle diseases related to dysfunctional mitochondria.

Keywords: Arthritis; Cellular organelle disease; Dysfunctional mitochondria; Inhibition of NOX4-p22phox axis; Mitocelle.

An intra articular injectable Mitocelle tracks dysfunctional mitochondria, and alleviates cartilage damage by targeting the NOX4-p22phox axis.

Fig. 1

Physicochemical properties of Mitocelle. (A) Schematic showing how Mitocelle can treat OA. (B) Photographs of dispersed and lyophilized Mitocelle. (C) Ultraviolet–visible absorption spectra, (D) hydrodynamic diameter and zeta potential, and (E) size distribution peaks of PB and Mitocelle. (F) Transmission electron microscopy images of PB (left) and Mitocelle (right). Scale bar, 500 nm. (G) X-ray diffraction spectra of PB and Mitocelle. (H) Immunofluorescence microscopy images of DAPI, ZO-1 (a membrane marker), and Cy5.5-labeled Mitocelle (left). Scale bar, 5 μm. Quantification of the fluorescence intensity of Cy5.5 (right). Data are presented as means ± SD (n = 5) and those in (H) were assessed using the Mann-Whitney U test. ∗∗P < 0.01.

Fig. 2

Mitocelle can track and bind stressed mitochondria. (A) Schematic diagram of the human protein microarray analysis. (B) Mitocelle-interacting proteins were analyzed using the HuProt™ 3.1 human protein chip. The signal-to-noise ratio (SNR) for each spot was calculated as the ratio of the foreground-to-background signal. In addition, the GST signal intensity (red) was used for SNR normalization (left). A high-power image of NOX4 binding (white circles) is shown in the right panel. (C) Chord diagram visualizing the relationship between proteins with SNR > 1.0 and a list of mitochondrial, Golgi, and ER proteins. The SNR was calculated using the formula SNR = 20 log₁₀ (Is In⁻¹), where ‘Is’ indicates the signal and ‘In’ indicates the noise. (D) Mouse chondrocytes treated with and without H₂O₂ were analyzed by real-time live imaging of mitochondria (green) and Mitocelle (red) using a Celldiscoverer7 and an LSM900 confocal microscope. Intensity profiles of linear regions of interest are shown in the right panel. (E) To confirm mitochondrial dysfunction, we performed JC-1 staining and quantified JC-1 aggregates (red) and JC-1 monomers (green) as average intensities expressed in arbitrary units. Data are presented as means ± SD (n = 5) and were assessed using one-way ANOVA with Bonferroni's test. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

Fig. 3

NOX4 expression is altered in osteoarthritis (OA). (A) NOX4 expression in human OA patient cartilage (left) and relative densitometry (right). Scale bar, 100 μm. (B) Expression of NOX4 in cartilage from mice with DMM-induced OA (left) and relative densitometry (right). Scale bar, 100 μm. (C) Gene set enrichment analysis (GSEA) of signature genes in chondrocytes infected with Ad-C or Ad-NOX4. (D) Expression levels of NOX4, MMP3, MMP13, and COX2 proteins in Ad-NOX4-infected chondrocytes, as detected by Western blot analysis (left) and analyzed by relative densitometry (right). (E) Intracellular ROS levels measured using DCF-DA in chondrocytes infected with Ad-NOX4 (800 MOI). (F) Safranin-O staining and immunohistochemistry of NOX4 and 8-OHdG (ROS marker) in DMM-induced cartilage of wild-type and NOX4^−/− mice. Scale bar, 100 μm. (G) Cartilage degradation was evaluated by OASRI score, osteophyte formation, and subchondral bone plate thickness (SBP). (H) Relative density of NOX4 and 8-OHdG signals in DMM-induced cartilage of wild-type and NOX4^−/− mice. Yellow dotted lines indicate tidemarks (F). Data are presented as means ± SD (n = 5) and were assessed using (A, B, E, G, H) Mann-Whitney U test or (D) one-way ANOVA with Bonferroni's test. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001.

Fig. 4

Enhanced penetration and extended retention of Mitocelle in cartilage matrix. (A) Fluorescence microscopy images of mouse articular cartilage explants exposed to free Cy5.5 or Cy5.5-labeled Mitocelle for 24 h (left). Quantification of the fluorescence penetration depth (right). Scale bar, 100 μm. (B) Fluorescence microscopy images of mouse articular cartilage sampled at 24 h after intra-articular injection of Cy5.5-labeled Mitocelle (red). Scale bar, 20 μm. Quantification of the fluorescence intensity of Cy5.5 (right). (C) Retention time of free Cy5.5 and Cy5.5-labeled Mitocelle injected into the knee joints of C57BL/6 mice (left). Data are presented as means ± SD (n = 5) and were assessed using (A, B) Mann-Whitney U test or (C) one-way ANOVA with Bonferroni's test. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001.

Fig. 5

Mitocelle alleviates cartilage destruction by downregulating ROS levels. (A) GSEA of Ad-NOX4-infected chondrocytes treated with or without Mitocelle. (B) Chondrocytes infected with Ad-NOX4 (800 MOI) were treated with Mitocelle, and intracellular ROS levels were assessed using DCF-DA. (C) Safranin-O staining and (D) immunohistochemical analysis examining the effect of Mitocelle in DMM-induced mice. After DMM surgery, intra-articular injection of Mitocelle was performed at 4-week intervals. Scale bar, 100 μm. (E) To evaluate the degree of cartilage damage, OARSI grade, osteophyte formation, and subchondral bone plate thickness were measured. (F) Relative densitometry assessing the expression levels of the OA-related catabolic factors, MMP3, MMP13, COX2, and 8-OHdG. Yellow dotted lines indicate tidemarks (C). Data are presented as means ± SD (n = 5) and were assessed using (B) one-way ANOVA with Bonferroni's test or (E, F) Mann-Whitney U test. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗∗P < 0.0001.

Fig. 6

Mitocelle interferes with the NOX4-p22phox interaction that contributes to ROS generation. (A) HuProt™ 3.1 Human Protein chip was used to analyze the effects of free Cy5 and Cy5-labeled Mitocelle (5 μg/mL) on Mitocelle-interacting proteins. The signal-to-noise ratio (SNR) of each point was calculated as the ratio of foreground to background signal. Additionally, a high-power image of p22phox binding (white circle) is shown in the right panel. (B) The SNR >1.0 proteins included 10 proteins known to be involved in ROS generation. (C) Schematic illustration of the principle of FRET and the acceptor photobleaching used for FRET measurements; CFP (donor), YFP (acceptor). (D) FRET detection by acceptor photobleaching. Chondrocytes were transfected for 24 h with vectors encoding YFP-NOX4 and CFP-p22phox, Mitocelles were applied, and transfection was continued for an additional 24 h. Fluorescence images were collected using YFP and CFP channels before and after photobleaching. To better show the changes in CFP fluorescence, pre- and post-bleaching CFP images are presented using pseudocolor. (E) FRET efficiency was measured after acceptor bleaching. Data are presented as means ± SD (n = 10) and were assessed using (E) Mann-Whitney U test. ∗∗∗∗P < 0.0001.