Nasal administration of mitochondria reverses chemotherapy-induced cognitive deficits

Abstract

Up to seventy-five percent of patients treated for cancer suffer from cognitive deficits which can persist for months to decades, severely impairing quality of life. Although the number of cancer survivors is increasing tremendously, no efficacious interventions exist. Cisplatin, most commonly employed for solid tumors, leads to cognitive impairment including deficits in memory and executive functioning. We recently proposed deficient neuronal mitochondrial function as its underlying mechanism. We hypothesized nasal administration of mitochondria isolated from human mesenchymal stem cells to mice, can reverse cisplatin-induced cognitive deficits. Methods: Puzzle box, novel object place recognition and Y-maze tests were used to assess the cognitive function of mice. Immunofluorescence and high-resolution confocal microscopy were employed to trace the nasally delivered mitochondria and evaluate their effect on synaptic loss. Black Gold II immunostaining was used to determine myelin integrity. Transmission electron microscopy helped determine mitochondrial and membrane integrity of brain synaptosomes. RNA-sequencing was performed to analyse the hippocampal transcriptome. Results: Two nasal administrations of mitochondria isolated from human mesenchymal stem cells to mice, restored executive functioning, working and spatial memory. Confocal imaging revealed nasally delivered mitochondria rapidly arrived in the meninges where they were readily internalized by macrophages. The administered mitochondria also accessed the rostral migratory stream and various other brain regions including the hippocampus where they colocalized with GFAP⁺ cells. The restoration of cognitive function was associated with structural repair of myelin in the cingulate cortex and synaptic loss in the hippocampus. Nasal mitochondrial donation also reversed the underlying synaptosomal mitochondrial defects. Moreover, transcriptome analysis by RNA-sequencing showed reversal of cisplatin-induced changes in the expression of about seven hundred genes in the hippocampus. Pathway analysis identified Nrf2-mediated response as the top canonical pathway. Conclusion: Our results provide key evidence on the therapeutic potential of isolated mitochondria - restoring both brain structure and function, their capability to enter brain meninges and parenchyma upon nasal delivery and undergo rapid cellular internalization and alter the hippocampal transcriptome. Our data identify nasal administration of mitochondria as an effective strategy for reversing chemotherapy-induced cognitive deficits and restoring brain health, providing promise for the growing population of both adult and pediatric cancer survivors.

Keywords: chemobrain; mesenchymal stem cell; mitochondria; nasal delivery; nrf2.

Figure 1

Nasal administration of mitochondria derived from human MSC resolves cisplatin-impaired executive function. (A) Schematic representation of cisplatin-treatment regimen, mitochondria-based therapeutic strategy and cognitive testing. Mice were intraperitoneally injected with cisplatin at 2.3 mg/kg for 5 consecutive days, followed by 5 days of rest and another 5 days of cisplatin injection. On the second and fourth day following the last cisplatin injection, mitochondria were freshly isolated from human MSC and administered intranasally. 14 days later, their cognitive behavior was assessed. (B) Schematic representation of PBT for evaluating executive function. Mice in the bright chamber were subject to access the dark chamber at three levels of complexity - an open tunnel (easy trials 1-4), bedding-covered tunnel (intermediate trials 5-7) and plugged tunnel (difficult trials 8-11). (C) The time taken to enter the dark chamber was measured. Cisplatin-treated mice were either slow or failed to unplug the tunnel entry in the difficult trials. Mice, nasally administered with mitochondria performed efficiently (n = 16-22). (D) The restoration of the executive function was determined at 34, 100 and 340 µg of mitochondrial protein with significant effects at 340 µg, in the difficult trials (n = 4-22). Cisplatin-impaired executive function was resolved by nasal administration of mitochondria in both (E) male (n = 16-22) and (F) female mice (n = 6) as seen in the mean time taken to remove the plug and enter the dark chamber. Results are expressed as mean ± SEM; Two-way ANOVA with Tukey's post hoc analysis *p ≤ 0.05; **p ≤ 0.01; **** p ≤ 0.0001.

Figure 2

Nasal administration of mitochondria resolves cisplatin-impaired working and spatial memory. (A) Schematic representation of NOPRT for evaluating spatial and working memory of mice. Mice were introduced to 2 identical objects during the training period, subsequently returned to their cages and then replaced in the testing arena with the familiar object at the original location and a new object at the opposite corner. (B) The mice movement and time spent with each object were tracked. The discrimination index was calculated as (T_Novel - T_Familiar) / (T_Novel + T_Familiar). Cisplatin-treated mice showed almost no preference for the novel object whereas those nasally administered with mitochondria performed comparable to the healthy control mice (C) The total interaction time did not differ between PBS/Vehicle and Cisplatin/Vehicle groups (n = 24-28) (D) Schematic representation of Y-maze test for assessing spatial memory. Mice were placed in one arm and their movement was monitored. (E) Perfect alternation was calculated from their sequential entry into all the arms before revisiting an arm. Cisplatin-treated mice showed decreased perfect alternations while mitochondrial administration reversed this impairment. (F) No difference was observed in the total number of arm entries between the different treatment groups (n = 18-22). Results are expressed as mean ± SEM; n = 18-28; Two-way ANOVA with Tukey's post hoc analysis *p ≤ 0.05; **p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001.

Figure 3

Nasally administered mitochondria arrive in the meninges. (A) Representation of meninges whole-mount with ROIs depicting where mitochondria were detected; scale bar 2500 µm. DsRed⁺/anti-human mitochondria⁺ mitochondria (yellow) were detected 30 min after administration near the meningeal (B) olfactory and (C) sinus regions; scale bar 100 µm. The mitochondria were in close contact with different cell types including (D) F4/80⁺ (magenta) and CD45⁺ (cyan) cells; scale bar 50 µm. 3D orthogonal slice views indicate the mitochondria are within the F4/80⁺ (magenta, D') and CD45⁺ (cyan, D'') cell bodies; DAPI⁺ nuclei appear blue. The orthogonal slice view shows the mitochondria (appearing yellow) are either well within the cell body or close to the nucleus (perinuclear); scale bar 10 µm. We did not detect human mitochondria internalized by (E) Lyve-1 (magenta) cells; scale bar 25 µm. 3D orthogonal slice view indicates the mitochondrion is outside the Lyve-1⁺ (magenta, E') cell body; scale bar 10 µm. (F) Nasally delivered DsRed⁺ (green)/anti-human mitochondria⁺ (red) mitochondria (yellow) predominantly interacted with F4/80⁺ meningeal macrophages (magenta); scale bar 25 µm and were internalized by these cells within 30 min of administration. The human mitochondria were localized within the cell body or in the perinuclear region as shown by 3D orthogonal slice view; scale bar 10 µm. Images taken with 20, 40 and 63x objectives.

Figure 4

Nasally administered mitochondria enter the brain. (A) Representation of mouse brain sagittal section with ROIs depicting where mitochondria were detected; scale bar 2500 µm. Nasally administered DsRed⁺/anti-human mitochondria⁺ mitochondria (yellow) were detected in (B) Ventricle; scale bar 100 µm (C) RMS - DsRed⁺ (green)/anti-human mitochondria⁺ (red) mitochondria (yellow) internalized by GFAP⁺ cells (magenta) in the RMS; scale bars 25 and 10 µm (D) Choroid Plexus (E) Hippocampus (F) Olfactory bulb; scale bars 25 and 50 µm.

Figure 5

Nasally administered mitochondria enter the pia mater and glial limitans. (A) Pia mater with F4/80⁺ cells (magenta) adhered to the brain with administered human mitochondria (yellow); scale bar 50 µm (B) internalized by cells in the pia mater and GFAP⁺ cells in the glial limitans (magenta) as indicated by (B' and B”) 3D orthogonal slice views; scale bar 25 µm and (C) the brain parenchyma; scale bar 50 µm. Images taken with 20, 40 and 100x objectives.

Figure 6

Nasally administered mitochondria repair cisplatin-damaged white matter integrity. (A) Mouse brain coronal section stained with Black Gold II with ROI indicating the cingulate cortex where white matter integrity was analysed; scale bar 1000 µm. Cisplatin-treatment (B) reduced percent area stained with Black Gold II and (C) increased coherency index which inversely relates to the complexity of myelin organization. Nasal administration of mitochondria reversed this damage. (D) Low magnification (4x) images representing Black Gold II⁺ area; scale bar 100 µm. (E) Higher magnification images (20x) representing the myelin organization; scale bar 25 µm. (F) Skeletonization of these higher magnification images revealed the complexity of the myelin arborization; scale bar 25 µm. Results are expressed as mean ± SEM; n = 4-8; Two-way ANOVA with Tukey's post hoc analysis **p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001.

Figure 7

Nasally administered mitochondria reverse cisplatin-induced synaptic loss. (A) Mice hippocampal regions stained with synaptophysin for different treatment groups; scale bar 25 µm. ROI indicates (B) clear synaptophysin⁺ synaptic puncta at higher magnification; scale bar 5 µm. Cisplatin-treated mice showed reduction in (C) percentage of synaptic puncta and (D) sum intensity while nasal administration of mitochondria reversed this loss. Results are expressed as mean ± SEM; n = 8; Two-way ANOVA with Tukey's post hoc analysis *p ≤ 0.05; **p ≤ 0.01; *** p ≤ 0.001.

Figure 8

Nasally administered mitochondria resolve cisplatin-induced synaptosomal mitochondrial defects and membrane integrity. (A) Representative TEM images depicting the ultrastructure of mitochondria in brain synaptosomes for different treatment groups. Mitochondrial swelling, membrane ruffling and cristae disorganization indicate defective or atypical mitochondria; scale bar 100 nm. (B) Representative TEM images showing whole synaptosomes. Membrane ruffling and disruption, blebs and vesicle leakage were considered to indicate compromised membrane integrity; scale bar 100 nm. Synaptosomes of cisplatin-treated mice revealed (C) high percentage of atypical mitochondria and (D) high percentage of damaged synaptic membrane. Results are expressed as mean ± SEM; n = 4; Two-way ANOVA with Tukey's post hoc analysis *p ≤ 0.05; **p ≤ 0.01.

Figure 9

Nasally delivered mitochondria induce transcriptomic changes in the hippocampus, the brain region crucial for cognition. RNA-sequencing was performed on hippocampi collected 72 h after the last administration of mitochondria. (A) Cisplatin treatment revealed 14874 stably expressed genes of which 1813 genes were systematically differentially expressed with fold change < 1/1.2 and > 1.2. (B) Nasal administration of mitochondria to cisplatin-treated mice revealed 14736 stably expressed genes of which 1308 were systematically differentially expressed. n = 4.

Figure 10

Nasally delivered mitochondria reverse cisplatin-altered genes in the hippocampus. RNA-sequencing was performed on hippocampi collected 72 h after the last administration of mitochondria. (A) Nasal administration of mitochondria reversed 698 genes altered by cisplatin-treatment in the hippocampus. Co-expression matrix shows subsets of genes decreased by cisplatin-treatment and increased by mitochondria as well as those increased by cisplatin and decreased by mitochondria. (B) Heat map of gene expression based on standardized z scores reveal expression pattern across different treatment conditions. Top 15 correlating genes based on strictly standardized mean difference (C) downregulated by cisplatin and upregulated by mitochondria and (D) upregulated by cisplatin and downregulated by mitochondria. n = 4.

Figure 11

Nasally delivered mitochondria activate key canonical pathways and regulate crucial functions in restoring cisplatin-induced cognitive deficits. Ingenuity Pathway Analysis of stably expressed genes revealed (A) top relevant canonical pathways activated and (B) crucial functions regulated by the nasal administration of mitochondria in cisplatin-treated mice. n = 4.