The MFN2 Q367H variant from a patient with late-onset distal myopathy reveals a novel pathomechanism connected to mtDNA-mediated inflammation
=============================================================================================================================================

* Mashiat Zaman
* Govinda Sharma
* Walaa Almutawa
* Tyler GB Soule
* Rasha Sabouny
* Matt Joel
* Gerald Pfeffer
* Timothy E. Shutt

## Abstract

**Background** *MFN2* encodes a multifunctional mitochondrial protein best known for its role mitochondrial fusion. While pathogenic variants in *MFN2* typically cause Charcot-Marie-Tooth disease subtype 2A, an axonal peripheral neuropathy, exome sequencing identified an uncharacterized *MFN2* variant, Q367H, in a patient diagnosed with late-onset distal myopathy without peripheral neuropathy. Although impaired mitochondrial fusion can cause mtDNA-mediated inflammation via TLR9 activation of NF-kB, which is linked to myopathy in a mouse model of MFN1 deficiency, this pathway has not yet been functionally linked to *MFN2* pathology.

**Methods** To investigate if the Q367H MFN2 variant contributes to the patient phenotype, we applied several biochemical and molecular biology techniques to characterize patient fibroblasts and transdifferentiated myoblasts for several functions mediated by MFN2. We also examined TLR9 and cGAS-STING mtDNA-mediated inflammatory pathways.

**Findings** Patient fibroblasts showed changes consistent with impairment of several MFN2 functions. When grown in standard glucose media, patient fibroblasts had reduced oxidative phosphorylation and elevated levels of lipid droplets. When grown in galactose media, patient fibroblasts had fragmented mitochondria, reduced mito-ER contact sites, and enlarged mtDNA nucleoids. Notably, under both media conditions, mtDNA was present outside of the mitochondrial network, where it co-localized with early endosomes. We also observed activation of both TLR9/NF-kB and cGAS-STING inflammation in fibroblasts. Moreover, the inflammatory signaling was increased 3-10 fold in transdifferentiated patient myoblasts, which also exhibited reduced mito-ER contacts and altered mtDNA nucleoids.

**Interpretation** We report a patient with myopathy, but without the typical peripheral neuropathy associated with *MFN2* disease variants. As elevated inflammation can cause myopathy, linking the Q367H MFN2 variant with elevated TLR9 and cGAS/STING signaling, which is amplified in transdifferentiated myoblasts, provides novel insight into the patient’s phenotype. Thus, we establish a potential novel pathomechanism connecting MFN2 dysfunction to mtDNA-mediated inflammation.

## Introduction

Mitochondria are highly dynamic cellular structures that fuse, divide, move about the cell, and interact with other organelles. The integral outer-mitochondrial-membrane (OMM) proteins Mitofusin1 (MFN1) and Mitofusin 2 (MFN2) mediate fusion of the outer mitochondrial membrane. MFN2, though named for its role in mitochondrial fusion, is also actively involved in several other dynamic processes (1): it mediates contacts between mitochondria and the endoplasmic reticulum (ER) (2), referred to here as mito-ER contact sites (MERCs); is involved in mitochondrial interactions with lipid droplets (3), early endosomes (4), and the nucleus (5); and also plays an active role in mitochondrial autophagy (mitophagy) (6,7).

Pathogenic variants in MFN2 are typically associated with the axonal peripheral neuropathy Charcot-Marie-Tooth disease subtype 2A (CMT2A). Characterized by weakness, numbness and pain in the distal regions of the body, CMT2A is often an early-onset progressively-worsening disease (8). In addition to peripheral neuropathy, patients with certain MFN2 variants can present with other pathogenic phenotypes, such as myopathy (9–11), optic atrophy, sensorineural hearing loss, ataxia (12), or multiple symmetric lipomatosis (MSL) (13–15). How and why different variants in MFN2 lead mechanistically to pathology is not fully understood, though it is likely not as simple as reduced mitochondrial fusion (16).

In addition to its critical role in neurons, MFN2 is also important for muscle function. Muscle-specific MFN2 knockout results in muscle atrophy, which is attributed to loss of mitochondrial fusion and subsequent instability of the mitochondrial genome (mtDNA) (17). On top of the depletion of mtDNA copy number that is thought to contribute to respiratory deficiency, mtDNA can accumulate point mutations and deletions in the muscle tissue of these KO mice. A role for MFN2 in maintaining mtDNA is further supported by the fact that certain pathogenic *MFN2* variants show a reduction in mtDNA copy number (18,19) or lead to mtDNA deletions (20). However, highlighting our incomplete understanding of the pathology caused by MFN2 deficiency and the role played by mtDNA, not all pathogenic *MFN2* variants impair mitochondrial fusion, impact mtDNA, or cause respiratory deficiency (16).

Beyond its role of encoding OXPHOS proteins, mtDNA is now also recognized to play an important role as an inflammatory signalling molecule when released from mitochondria. Examples of inflammatory signalling pathways that detect mtDNA-mediated include TLR9 (4,18), cGAS-STING (21,22) and the inflammasome (23,24). Notably, a recent study showed that inducible muscle-specific loss of MFN1 causes mtDNA release via early endosomes, leading to activation of TLR9/NF-kB inflammation and myopathy (4). Importantly, pharmacological inhibition of this inflammation improved the muscle atrophy and physical performance in these mice, implicating sterile inflammation in the muscle pathology of these mice. Nevertheless, the role of mtDNA-mediated inflammation in the pathology of pathogenic *MFN2* variants has not been described to date.

Here we report a novel MFN2 variant, Q367H, in a patient diagnosed with late onset distal myopathy. Functional characterization of patient fibroblasts and transdifferentiated myoblasts reveals mtDNA-mediated TLR9 signaling that likely explain the patient myopathy, thus expanding our understanding of the mechanisms by which MFN2 dysfunction contributes to pathology.

## Methods

### Ethics statement

The patient provided written informed consent for participation in research and consented to the publication of this report. He was provided with a copy of the draft article prior to submission. Genetic sequencing and human tissue studies with this participant were part of research protocols (REB15-2763 and REB17-0850) approved by the Conjoint Health Research Ethics Board at the University of Calgary.

### Exome Sequencing, Variant Confirmation and Sequence Conservation Study

DNA was extracted from blood collected into EDTA tubes using standard protocols. Library preparation proceeded using the Ion Ampliseq Exome RDY Panel (Thermo Fisher, A38264) according to manufacturer’s protocol. Automated chip loading and templating used the Ion Chef system and 540 chip/chef kit (Thermo Fisher, A30011), and sequencing was performed on an Ion S5 system (Thermo Fisher, A27212), according to manufacturer’s protocols. Base calling, read alignment to hg19, coverage analysis, and variant calling were performed with Torrent Suite (v. 5.10.1; Thermo Fisher). Similar analysis can be performed using an alternative bioinformatic pipeline as described previously (25). Patient VCFs were annotated for predicted variant consequence, gnomAD allele frequency, CADD score, and OMIM phenotypes, in addition to default parameters with Ensembl’s command line Variant Effect Predictor (VEP). We additionally aligned off-target reads from exome sequencing to identify any mtDNA variants, as previously described (26).

The variant was verified by Sanger sequencing. Briefly, total genomic DNA was collected using the E.Z.N.A Tissue DNA Extraction Kit (VWR, CA101319-018). The region of interest was amplified using PCR with the following primers (F-GAGAAGAGCAGAGAGGCTCTTG and R-ACACAGGGAATCGTGCTGC). Subsequently the amplified region was sequenced by Sanger sequencing at University of Calgary Centre For Health Genomics And Informatics.

Amino acid sequence alignment was performed using the T-COFFEE online tool (27). The sequences were collected from UniProt(28). The following accession codes were used for the sequences: MFN2: [O95140](http://medrxiv.org/lookup/external-ref?link\_type=GEN&access_num=O95140&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom)-1, MFN1: Q8IWA4, Mouse MFN2: Q80U63, Fish MFN2: A8WIN6, Chicken MFN2: E1BSH7, Frog MFN2: Q28CS2, Fruit fly MFN: Q7YU24, Worm MFN: [Q23424](http://medrxiv.org/lookup/external-ref?link_type=GEN&access_num=Q23424&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom).

### Quantification of Protein Amounts

Relative quantities were measured using Western Blot Analysis, as previously described (12). Briefly, cells were lysed with Pierce RIPA buffer (Thermo Scientific™, 89900), and subsequently total protein was quantified by a BCA assay. 50 μg of protein was loaded from control and patient fibroblasts on to an SDS-PAGE gel. After running the gel, the blots were transferred to a PVDF membrane overnight. The blots were probed using the following antibodies: anti-MFN2 (Abnova, H00009927-M03; 1:1000), anti-MFN1 (Cell Signaling Technology, D6E2S; 1:1000), anti-OPA1 (BD Transduction Laboratories, 612607; 1:1000), anti-Actin (Sigma, A5316; 1:1000). Secondary antibodies complimentary to the primary host species were used-goat anti-rabbit IgG, HRP linked Antibody (Cell Signaling Technology, 7074S) or goat anti-mouse IgG-HRP (Santa Cruz Biotechnology, sc-2055). The membranes were visualized with SuperSignal™ West Femto Maximum Sensitivity Substrate (Thermo Scientific™, 34095) using the Bio-Rad ChemiDoc Imaging System. Bands were quantified using Fiji (29).

### Cell Culture and Maintenance

Fibroblasts from the patient and healthy individual were generated through skin biopsies as described (25). The cells were cultured using Minimum Essential Media (Gibco, 11095080), supplemented with 10% Fetal Bovine Serum and 1mM Sodium Pyruvate. Glucose-free conditions were established using DMEM No Glucose media (Gibco, 11966025), supplemented with 5mM galactose, 10% FBS and 1mM Sodium Pyruvate. Cells were grown in an incubator, maintained at 37°C and 5% CO2. For the fatty acid pulse-chase assay, the cells were grown in the glucose media mentioned above, or glucose-free media with 10% FBS, but no galactose.

### Mitochondrial Morphology Analysis

Mitochondrial morphology was analyzed by immunostaining as described previously (30), where 30,000 cells were grown in 12mm coverslips for 24 hours and subsequently fixed by 4% PFA for 20 minutes at 37°C. Cells were permeabilized using 0.1% Triton X-100 (MP Biomedicals, J00105) and blocked using 5% FBS. The OMM protein TOMM20 was chosen to visualize the mitochondrial network, where cells were incubated with mouse anti-TOMM20 (Santa Cruz Biotechnology, F-10, RRID: AB_628381) overnight, the subsequently probed with goat anti-mouse secondary conjugated with Alexa fluor 488 (Thermo Fisher Scientific, A-11029). Cover slips were then mounted on glass slides using ProLong Diamond Antifade Mountant with DAPI (Invitrogen, P36966).

### Oxygen Consumption Rate

The cellular oxygen consumption rate assay for control and patient fibroblasts was performed as previously described (31). Briefly, 40,000 cells were seeded in wells of the XF-24 plates (Agilent). The media on the plates was replaced 24 hours later with Seahorse CF Base Medium, supplemented with 1mM Pyruvate, 2mM Glutamine and 10mM Glucose and brought to pH 7.4. Post-incubation for 45 minutes with CO2, cells were loaded into the Seahorse XF24 Analyzer. Oxygen consumption rates were measured at three time points, following administration of Oligomycin (0.5 μM), Carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP) (1 μM) and Antimycin A (0.5 μM). Following the experiment, cells in each well were lysed with Pierce RIPA buffer and total protein for each well was quantified with BCA assays, for normalization purposes.

### mtDNA and Mitochondrial Nucleoid Analyses

Mitochondrial nucleoids were visualized using confocal microscopy as previously reported (32). Briefly, 30,000 control and patient fibroblasts were grown on coverslips for 24 hours and fixed with 4% PFA. Cells were permeabilized with 0.2% Triton X-100 and blocked with 5% FBS. Mitochondrial nucleoids were labelled using mouse anti-dsDNA antibody (Developmental Studies Hybridoma Bank, AB_10805293, RRID:AB_10805293) and subsequently visualized using the goat anti-mouse secondary conjugated with Alexa fluor 488. The mitochondrial network was also labelled using anti-rabbit TOMM20 rabbit anti-TOMM20 (Abcam, ab186735), which was visualized with anti-Rabbit secondary conjugated with Alexa fluor 647 (Thermo Fisher Scientific, A-21245). Cover slips were mounted using ProLong Diamond Antifade Mountant with DAPI.

mtDNA copy number analysis was performed using qPCR. Total DNA was collected from control and patient fibroblasts seeded at the same density. Relative mtDNA copy number was analyzed using nuclear-encoded housekeeping gene 18S, by amplification using the following primers:

mito-F-CACCCAAGAACAGGGTTTGT

mito-R-TGGCCATGGGTATGTTGTTAA

18S-F-TAGAGGGACAAGTGGCGTTC

18S-R-CGCTGAGCCAGTCAGTGT

For qPCR, 50ng of total DNA was loaded into each well a 96-well plate and the reaction was performed using the QuantStudio6 Flex Real-Time PCR system (Thermo Fisher Scientific). The delta Ct method was used to determine the relative copy number.

Long range PCR was done to look for mtDNA deletions using the Takara LA Taq Polymerase (Takara Bio, RR002M). Total DNA was collected from control and patient fibroblasts and 16.2kBP fragment was amplified using the following primers- (F-ACCGCCCGTCACCCTCCTCAAGTATACTTCAAAGG and R-ACCGCCAGGTCCTTTGAGTTTTAAGCTGTGGCTCG). Long Range PCR was done using the following conditions- (94 °C for 1 min; 98 °C for 10 s, and 68 °C for 11 min (30 cycles), followed by a final extension cycle at 72 °C for 10 min). The products following the reaction were visualized by gel electrophoresis on a 0.6% gel run for 18 hours at 18V.

### Mito-ER Contact Sites

The proximity ligation assay was used to look for MERCs, using the Duolink In Situ Detection kit (Sigma Aldrich, DUI92008). Briefly, 30,000 control and patient cells were grown on coverslips for 24 hours and subsequently fixed using 4%PFA. Cells were permeabilized with 0.1% Triton X-100 and blocked using the blocking reagent provided by the kit. The cells were then incubated overnight using mouse-anti-Calnexin (Sigma-Aldrich, MAB3126) and rabbit anti-TOMM20 (Abcam, ab186735). Following incubation, PLA probes, complementary to the primary antibodies, anti-mouse (Sigma-aldrich, DUO92004) and anti-rabbit (Sigma-aldrich, DUO92002) were added to the coverslips. The addition of DNA ligase and DNA polymerase were subsequently carried out using manufacturer’s instructions and using the reagents provided in the kit. Finally, secondary antibodies anti-mouse secondary conjugated with Alexa fluor 488 (Thermo Fisher Scientific, A-11029) and anti-Rabbit secondary conjugated with Alexa fluor 647 (Thermo Fisher Scientific, A-21245) were added for visualizing the ER and mitochondrial networks respectively. The coverslips were mounted with ProLong Diamond Antifade with DAPI before imaging.

### Lipid Droplets and Fatty Acid Pulse-Chase

Total cellular lipid droplets were labelled using HCS LipidTox Green (Thermo Fisher Scientific, H34350), following the same immunostaining protocol to label the mitochondrial network, except for using 0.1% Saponin (Sigma Aldrich, SAE0073-10G) as the permeabilizing solution. The mitochondrial network was also labelled using anti-rabbit TOMM20 rabbit anti-TOMM20 (Abcam, ab186735), which was visualized with anti-Rabbit secondary conjugated with Alexa fluor 647 (Thermo Fisher Scientific, A-21245).

For the fatty acid pulse-chase assay, 50,000 cells were seeded on a 35mm glass bottom dish with 20mm indented imaging region. Cells were grown for 24 hours, after which they were pulsed with the fluorescent fatty acid precursor BODIPY 558/568 C12 (Invitrogen, D3835). After 24 hours incubation with the BODIPY, cells were incubated with either glucose or glucose-free media and chased at time points 0 hours, 6 hours and 24 hours following the addition of the two media conditions. When the variable media conditions were added, MitoTracker Green (Invitrogen, M7514) was also added at 50nM concentration for visualizing the mitochondrial network.

### Mitochondria, mtDNA and Early Endosome Imaging

The triple staining of the mitochondrial network, mtDNA and early endosomes was performed using sequential immunostaining. The steps until blocking are the same as described above for the mitochondrial network. Following blocking, primary antibodies were added for the mtDNA (same as above) and mitochondrial network (same as above) and secondary antibodies were added accordingly-anti-mouse secondary conjugated with Alexa fluor 488 (Thermo Fisher Scientific, A-11029) and anti-Rabbit secondary conjugated with Alexa fluor 568 (Thermo Fisher Scientific, A-11004). Following secondary antibody incubation, coverslips were thoroughly washed with PBS and the primary antibody to label early endosomes was added rabbit anti-RAB5 (Cell Signaling Technology, 3547T). Subsequently, the complementary secondary antibody anti-Rabbit secondary conjugated with Alexa fluor 647 (Thermo Fisher Scientific, A-21245) was added. The coverslips were mounted using ProLong Diamond Antifade mounting media.

### Fibroblast Transdifferentiation

The coding sequence for human myoD1 (accession [NM\_002478](http://medrxiv.org/lookup/external-ref?link_type=GEN&access_num=NM_002478&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom)) was PCR amplified from a human skeletal muscle cDNA library and subcloned into pWPXL (pWPXL was a gift from Didier Trono (Addgene plasmid # 12257) in place of eGFP at the 5’BamHI and 3’EcoRI restriction sites. The DNA sequence was confirmed using Sanger DNA sequencing.

Lentivirus production was performed using 293FT cells grown to 90% confluency on five, 15 cm2 cell culture plates, by cotransfecting 112.5 μg pWPXL-myoD1, 73 μg psPAX2 and 39.5 μg pMD2.G [gifts from Didier Trono (Addgene, plasmid # 12260) using PEIpro transfection reagent (Polyplus). The cell culture media was collected 48 hours post-transfection, quantified using Lenti-XTM GoStixTM Plus (Takara Bio USA, 631280), centrifuged at 500xg for 5 min, 0.45 um filtered and stored at -80°C as 5 mL aliquots.

Non-immortalized primary fibroblasts below passage 5 were grown at 37°C, 5% CO2 in complete media: DMEM – high glucose (Corning, 10-013-CV), 10% FBS (Corning, 35-077-CV), 1% P/S (VWR, K952). T75s or 24 well plates were coated in 0.1 mg/mL VitroCol Type I human collagen (Advanced BioMatrix, 5008) for 1 hour before gently washing with PBS. Cells were seeded at 6,300/cm2 and transduced at 75% confluency. 5 µg/mL polybrene (Sigma-Aldrich, TR-1003-G) was mixed with MYOD1 lentivus at GoStix Values of either 11,500 or 12,000 depending on the biological replicate. Cells were transduced by spinfection, specifically a one-hour spin at 800 xg, RT, and then lentiviral supernatant was replaced with complete media. Media was changed every two days and cells were allowed to transdifferentiate for 7 days before fixing or RNA extraction. Conversion to myoblasts was confirmed by visualization of Desmin using Desmin (D93F5) XP Rabbit mAb (CST, cat. #5332S) and secondary AlexaFluor 594 chicken anti-rabbit (Invitrogen, A-21442). RNA for qPCR was extracted using RNeasy Mini Kit (QIAGEN, 74104).

### Analysis of TLR9-NF-kB Signalling Pathways

The expression of genes associated with the TLR9-NF-kB was analyzed by RT-qPCR. The TLR9 inhibitor Chloroquine (MilliporeSigma, C973P63) and the cGAS-STING inhibitor Ru.521 (Aobious, AOB37877) were added to the cell culture media (final concentration 1uM) for 24 hours, and subsequently the cells were washed five times with PBS. Briefly, total RNA was extracted from cells using the HP Total RNA Extraction Kit (VWR, CA101414-852), according to the manufacturer’s instructions. The collected RNA was reverse transcribed using the iScript Advanced cDNA Synthesis Kit (BioRad, 1725038). Subsequently, qPCR was performed using the QuantStudio 6 Real-Time PCR system (Thermo Scientific). The primers used for amplification of target genes are:

Beta Actin- F- AAGACCTGTACGCCAACACA

Beta Actin- R- AGTACTTGCGCTCAGGAGGA

Asc-F- CATGAACTGATCGACAGGATG

Asc-R- GGACCTCCTCCAAATGTTTC

IL6-F-- AGACAGCCACTCACCTCTTCAG

IL6-R- TTCTGCCAGTGCCTCTTTGCTG

Nlrp3-F- GCTGGCATCTGGATGAGGAA

Nlrp3-R- GTGTGTCCTGAGCCATGGAA

S100a9-F- GGAATTCAAAGAGCTGGTGC

S100a9-R- TCAGCATGATGAACTCCTCG

TNF-F- GCCCATGTTGTAGCAAACCC

TNF-R- GGAGGTTGACCTTGGTCTGG

IFN-α-F- AAATACAGCCCTTGTGCCTGG

IFN-α-R- GGTGAGCTGGCATACGAATCA

IFN-β-F- AAGGCCAAGGAGTACAGTC

IFN-β-R- ATCTTCAGTTTCGGAGGTAA

### Imaging and Analyses

All imaging was performed as previously described (21). The Olympus Spinning Disk Confocal System (Olympus SD OSR) was used with the UAPON 100XOTIRF/1.49 oil objective lens. For live cell imaging, a cellVivo incubation module was used to maintain cells at 37°C and 5% CO2. For analysis of mitochondrial branch length and footprint, the MiNA plugin (33) was used on Fiji. Number and size of PLA puncta, mitochondrial nucleoids and lipid droplets was performed using the particle analysis tool on Fiji, following thresholding. All co-localization studies were performed using the JaCoP plugin (34) on Fiji. All quantitative figures and statistical tests were performed using GraphPad Prism 9.

## Results

### Patient Report, Whole Exome Sequencing and Protein Expression for MFN2 Q367H Variant

A male patient in their 70s presented with a two-year history of slowly progressive weakness affecting his lower extremities. Initial symptoms included difficulty walking uphill, and he eventually developed difficulty with stairs and limitations with walking on the ground level and rising from a chair. There were no cramps, fasciculations, or contractures. No sensory loss or coordination. Dysarthria, dysphagia, ptosis, or diplopia were not present.

The patient was medically well and did not have any history of cardiac disease. The parent of the proband developed weakness in her lower legs in her 60s, with progressive limitations in her mobility. Apparently, she was diagnosed with CMT disease, but her clinical profile was not available for review. The parent died in her early 80s due to congestive heart failure. The parent’s two siblings aged 76-80, and her other two children aged 40-45 were all unaffected. Clinical examination of the patient demonstrated a cognitively normal individual. Cranial nerve examination was normal. Motor examination revealed MRC grade 4+ strength for ankle dorsiflexion and 4+ plantarflexion bilaterally. Proximal and axial muscle testing was normal, except for Achilles tendon reflexes. The sensory examination was normal. Creatine kinase enzyme activity was elevated to a maximum of 455 U/L (reference range ≤ 195 U/L). The muscle biopsy showed necrotic and regeneration myofiber, myofiber hypertrophy, multifocal endomysia, and perimysial fibrosis, and extensive infiltration by adipose tissue. Immunostaining for dystrophin, alpha/gamma sarcoglycan, sceptrin, α-laminin, desmin, and α-β-crystallin was unremarkable. There was no inflammatory cell infiltrate. For the magnetic resonance imaging (MRI) examination, upper and lower legs revealed atrophy in the adductor longus gluteus minimus, tibialis anterior, and inferior gastro-soleus bilaterally [Figure 1A/B].

![Figure 1:](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2024/06/21/2024.06.20.24309123/F1.medium.gif)

[Figure 1:](http://medrxiv.org/content/early/2024/06/21/2024.06.20.24309123/F1)

Figure 1: Initial characterization of patient phenotypes and MFN2 Q367H variant.
A. MRI image at the level of the legs demonstrating predominant atrophy in white arrows; Gastrocnemius (A), Soleus (B) and Tibialis Anterior (C) bilaterally. B. MRI image of upper Thighs demonstrating predominant atrophy of adductor longus gluteus minimus (Non shown here). C.(i) Sanger sequencing results showing single nucleotide polymorphism in patient fibroblasts, transition Guanine to Cytosine. (ii) Topology diagram of MFN2 protein with domains and localization indicated. The site of the Q367H variant is shown in the diagram. D. Amino acid sequence alignment using T-COFFEE tool. E. Western Blot analysis comparing expression of mitochondrial fusion protein in control vs patient fibroblasts, normalized to β-actin. (F-J) Comparison of protein expression levels (F) MFN2 (glucose media) (G) MFN2 (glucose-free/galactose supplemented media) (H) MFN1 (I) OPA1 levels in control vs patient fibroblasts. Data is indicative of mean +/- SD. Statistical comparisons were made by unpaired t-test, (ns=not significant (p>0.05)).

The case was interpreted as a chronic myopathy nonspecific muscle pathology. A genetic cause was suspected given the family history of a similar condition in his parent, and lack of inflammation on muscle biopsy. Clinical testing with a gene panel encompassing 200 genes associated with myopathy, neuropathy, and neuromuscular disorders identified a previously uncharacterized heterozygous variant in the *MFN2* gene, c.1101G>C, p. (Gln367His) [Table 1/Figure 1C]. Research based exome sequencing again identified the *MFN2* variant, as well as heterozygous variants of unknown significance in *SYNE1* and *DYSF* (Table 1). While pathogenic variants in *SYNE1* can be associated with autosomal dominant neuromuscular disease, this patient variant was not considered to be clinically relevant for the following reasons: (a) the variant is not in the same domain as pathogenic variants previously associated with autosomal dominant disease, (b) *SYNE1* variants in this region of the gene have only been associated with autosomal recessive disorders, while our patient is heterozygous, (c) no characteristic nuclear abnormalities associated with pathogenic variants in *SYNE1* (35) were observed in patient fibroblasts under microscopy, (d) the myopathy phenotype is not in keeping with the *SYNE1*-related disease phenotype, which includes proximal weakness and contractures. Meanwhile, the *DYSF* variant was not considered to be an explanation for this patient’s disease because it is a single heterozygous variant in a condition caused by biallelic variants in *DYSF*, and we are not aware of any reports of dysferlinopathy associated with only a single pathogenic variant. Thus, based on strong suspicion for the relationship between the *MFN2* variant and the patient phenotype, we proceeded with the studies described in this report using patient fibroblast cells.

View this table:
[Table 1.](http://medrxiv.org/content/early/2024/06/21/2024.06.20.24309123/T1)

Table 1. 

### Characterization of Patient Fibroblasts

To investigate if the Q367H variant contributes to the patient phenotype, we began by examining skin fibroblasts obtained from the patient. Immunoblotting for mitochondrial fusion proteins revealed no significant changes in the expression of MFN2, MFN1, or the inner mitochondrial membrane fusion protein OPA1 (Figure 1E-J). Next, to determine if this MFN2 variant may contribute to the pathology and to gain potential insight into the disease mechanism(s), we characterized patient fibroblasts for alterations to a variety of MFN2-mediated functions.

### Mitochondrial Network Morphology and Respiration in MFN2 Q367H Fibroblasts

First, as fibroblast cells from CMT2A patients with MFN2 variants sometimes have fragmented mitochondria (16), we examined mitochondrial morphology as a potential downstream readout of altered mitochondrial fusion (Figure 2A-C). No changes were observed under standard culture conditions. However, glucose-free growth conditions supplemented with galactose, which forces cells to use mitochondria for energy rather than glycolysis, can be used detect morphology changes in fibroblasts from patients with MFN2 variants (36). Consistent with decreased mitochondrial fusion and mitochondrial dysfunction, Q367H patient fibroblasts grown in glucose-free galactose-supplemented media showed a marked increase in the number of cells with fragmented mitochondrial networks, as well as a decrease in average mitochondrial length. Impairment of mitochondrial function in patient fibroblasts was further reinforced by functional studies of oxygen consumption. Patient fibroblasts had increased basal oxygen consumption but reduced maximal oxygen consumption (Figure 2D). As mitochondrial fragmentation alone does not indicate MFN2 dysfunction (*e.g.*, reduced mitochondrial health in general can lead to mitochondrial fragmentation), we also explored other functions performed by MFN2 to provide additional evidence for MFN2 dysfunction.

![Figure 2:](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2024/06/21/2024.06.20.24309123/F2.medium.gif)

[Figure 2:](http://medrxiv.org/content/early/2024/06/21/2024.06.20.24309123/F2)

Figure 2: Changes in mitochondrial morphology and respiratory function.
(A) Representative confocal images (top) with zoomed inset images (bottom) showing mitochondrial network morphology in control vs patient fibroblasts stained with TOMM20. (B) Qualitative scoring of proportion of cells showing fragmented/intermediate/fused mitochondrial network. Bars indicate mean +/-SD. (C) Quantitative analysis of mean mitochondrial branch length in control vs patient fibroblasts using glucose or glucose-free/galactose supplemented media, bars show mean +/- SD, colours indicate biological replicates. (D) Oxygen Consumption Rate (OCR) assay on control and patient fibroblasts using Seahorse XFe24, bars indicate mean +/- SD (n=12 replicates). Statistical comparisons were made by unpaired t-test (ns=not significant (p>0.05); \**\*|\*=p≤ 0.0001).

### Perturbated Mitochondrial Nucleoids in MFN2 Q367H Patient Fibroblasts

Mitochondria house their own genome, mtDNA, which is present in hundreds of copies throughout the mitochondrial network and is organized into nucleoid structures. As mtDNA nucleoids can be altered by impairments to mitochondrial fusion (37), we sought to examine mtDNA in Q367H patient fibroblasts (Figure 3A/B). Similar to mitochondrial morphology, there were no changes to mtDNA nucleoid size or abundance in standard media. However, under glucose-free galactose-supplemented conditions, patient fibroblasts had a decrease in the average number of mitochondrial nucleoids per cell, while the average nucleoid size was increased (Figure 3A/C). Of note, in both growth conditions, a small number of mtDNA nucleoids were observed outside the mitochondrial network in patient fibroblasts (circled in figure 3A). Meanwhile, the mtDNA copy number did not change under any conditions (Figure 3D/E), nor were any mtDNA deletions detected (Figure 3F). Observing fewer, larger nucleoids without changes in mtDNA copy number is likely due to nucleoid clustering, a phenotype often observed with impaired mitochondrial fusion (37). As such, the mtDNA studies are consistent with reduced fusion leading to mitochondrial fragmentation in the Q367H cells when grown in glucose-free galactose-supplemented media.

![Figure 3:](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2024/06/21/2024.06.20.24309123/F3.medium.gif)

[Figure 3:](http://medrxiv.org/content/early/2024/06/21/2024.06.20.24309123/F3)

Figure 3: Characterization of mtDNA nucleoids, mtDNA copy number, and mtDNA deletions.
(A) Representative confocal images showing mitochondria (TOMM20; grey) and mitochondrial nucleoids (dsDNA antibody; red) in patient and control fibroblasts under glucose and glucose-free/galactose supplemented nutrient conditions. Circles indicate presence of mtDNA nucleoids outside the mitochondrial network. (B) Number of nucleoids per cell, mean is shown with colours indicating biological replicates. (C) Average size of mitochondrial nucleoids, violin plots show median and interquartile ranges. (D-E) qPCR data showing mitochondrial DNA copy number in (D) glucose and (E) glucose-free/galactose supplemented media, bars indicate mean +/- SD. (F) Agarose gel electrophoresis showing long range PCR products used to detect full-length or deleted mtDNA and DNA ladder showing corresponding sizes. Statistical comparisons were made by unpaired t-test, (ns=not significant (p>0.05); *=p≤ 0.05; \***|=p≤ 0.001; \**\*|\*=p≤ 0.0001).

### MFN2 Q367H Fibroblasts show reduced number of Mito-ER Contact Sites

MFN2 also mediates mito-ER contact sites (MERCs) (38). As an estimate of MERCs, we used a proximity ligation assay to detect regions where the mitochondria and ER are within ∼40 nm. No absolute changes in the number or size of the PLA signals were observed in standard media. However, under glucose-free conditions, Q367H cells had fewer PLA puncta, though these puncta were slightly larger (Figure 4A-C). As changes in MERCs could be due to global changes in mitochondrial abundance, we also measured the area of mitochondria (mitochondrial footprint). Importantly, we do not observe significant differences between mitochondrial footprint in control and patient cells (Figure 4D). However, when the MERC PLA signal is normalized to the mitochondrial footprint, we see a slight a reduction in the number of MERCs in the patient fibroblasts in standard conditions, which is further exaggerated under glucose-free conditions (Figure 4E). Together, these data indicate perturbations in MERCs in the Q367H fibroblasts that are consistent with MFN2 dysfunction.

![Figure 4:](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2024/06/21/2024.06.20.24309123/F4.medium.gif)

[Figure 4:](http://medrxiv.org/content/early/2024/06/21/2024.06.20.24309123/F4)

Figure 4: Mito-ER Contact site alterations.
(A) Representative confocal images of mito-ER contact sites (MERCs) using proximity ligation assay (PLA) showing mitochondria (TOMM20; cyan), PLA probes (green) and ER (Calnexin; magenta) in glucose and glucose-free/galactose supplemented media. (B) Quantitative analysis of number of MERCs per cell, lines indicate mean +/- SD, colours indicate biological replicates. (C) Quantitative analysis of average size of PLA probes, with violin plots indicate median and interquartile range. (D) Quantitative analysis of mitochondrial area in each cell type in (μm²), lines show mean +/- SD, colours indicate biological replicates. (E) Normalized number of MERCs to total mitochondrial area per cell, lines indicate mean +/- SD, colours indicate biological replicates. P-values determined by unpaired t-test (ns=not significant (p>0.05); *=p≤ 0.05; **=p≤ 0.01; \**\*|\*=p≤ 0.0001).

### Dysfunction in Mitochondrial Fatty Acid Import Leads to Lipid Droplet Accumulation in Patient Fibroblasts

Given the role of MFN2 in mediating mitochondria-lipid droplet interactions, and the fact that lipid droplets are altered in fibroblasts from patients with MFN2 variants (16), we next explored lipid droplets in the Q367H fibroblasts. Interestingly, we see a significant increase in the number and size of cellular lipid droplets in patient fibroblasts, both grown in standard and glucose-free galactose-supplemented media (Figure 5A-D). To explore the mechanism underlying the increased accumulation of lipid droplets, we examined lipid transfer from lipid droplets to mitochondria using an assay that takes advantage of the BODIPY 558/568 fatty acid precursor (39). In this assay, the fluorescently labelled lipid is loaded into lipid droplets during a 24-hour incubation, and its transfer to mitochondria can be monitored over time by microscopy. In control cells in standard conditions, some fatty acid pre-cursor transfers into the mitochondrial network (Figure 5E-F), with higher transfer under glucose starvation conditions (without galactose supplementation). Notably, in patient fibroblasts, no fatty acid trafficking from lipid droplets to mitochondria is observed under either growth condition (Figure 5E-F). These findings suggest that the excess lipid droplets seen in the patient fibroblasts are due to an accumulation of lipids due to a lack of fatty acids transfer from lipid droplets to mitochondria.

![Figure 5:](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2024/06/21/2024.06.20.24309123/F5.medium.gif)

[Figure 5:](http://medrxiv.org/content/early/2024/06/21/2024.06.20.24309123/F5)

Figure 5: Changes to cellular lipid droplets.
(A-B) Representative confocal images showing lipid droplets stained with HCS LipidTox Green in (A) glucose and (B) glucose-free/galactose supplemented media. (C) Quantitative analysis of the average number of lipid droplets per cell, bars indicate mean +/- SD, colours indicate biological replicates. (D) Quantitative analysis of the average size of lipid droplets, violin plots show median and interquartile range. (E) Representative live-cell confocal images showing mitochondria (MitoTracker Green) and fatty acids (BODIPY 558/568) at 0 hours and 24 hours in standard or glucose-starved media. (F) Quantitative analysis of co-localization between fatty acid and mitochondrial network, colours indicate cell type (red-control and blue-MFN2 Q367H), bars indicate mean +/- SD. P-values determined by unpaired t-test (ns=not significant (p>0.05); **=p≤ 0.01; \**\*|\*=p≤ 0.0001).

### MFN2 Q367H Patient Cells Show mtDNA-Mediated Inflammation

Following up on observation of nucleoids outside of the mitochondrial network (Fig 3), we looked for evidence of mtDNA-mediated inflammation, which is linked to impaired mitochondrial fusion and can cause myopathy (4). First, to look for signs for an inflammatory response, we examined the expression of several inflammation genes via q-RT-PCR. Notably, even when cells were grown in standard glucose conditions, we observed an elevation of genes that mediate TLR9 signalling (*i.e.*, ASC/NLRP3/S100A9) and downstream cytokines of the TLR9 pathway (*i.e.*, IL6/TNFα), as well as genes downstream of the cGAS-STING pathway (INF-α/β) (Figure 6A). Similar inflammatory signalling was observed when fibroblasts were grown in galactose media (Supplemental Figure 1). To determine if the increased expression of these genes was due to the TLR9 or cGAS-STING signalling pathways, fibroblasts in standard media were treated with chloroquine to inhibit TLR9 signalling (40), or RU.521 (41) to inhibit cGAS (Figure 6A). Strikingly, chloroquine reduced TLR9 targets, while RU.521 reduced cGAS-STING targets. The fact that each treatment significantly reduced expression of the relevant genes to the levels seen in the control patient fibroblasts, confirms that both TLR9-NF-kB and cGAS-STING signalling are elevated in the patient fibroblasts and that these are the primary pathways driving expression.

![Figure 6:](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2024/06/21/2024.06.20.24309123/F6.medium.gif)

[Figure 6:](http://medrxiv.org/content/early/2024/06/21/2024.06.20.24309123/F6)

Figure 6: MFN2 Q367H cells show activation of TLR9 and cGAS-STING inflammatory pathways, and co-localization of mtDNA with early endosomes.
(A) RT-qPCR data showing gene expression for TLR9/NF-KB targets (Asc, IL6, NLRP3, S100a9, TNF) and cGAS-STING targets (Ifn-α, Ifn-β) in control vs patient fibroblasts under glucose conditions. The graph also shows the effects of inhibitors for the TLR pathway (Chloroquine) or cGAS-STING pathway (RU.521). Lines indicate mean +/- SD. (B) Representative confocal images showing mitochondria (TOMM20; cyan), mitochondrial nucleoids (anti-dsDNA antibody; green), and early endosomes (RAB5C;magenta). Images in inset bottom right panel highlights the location of mitochondrial nucleoids outside the mitochondria with the early endosomes, denoted by the white arrows. (C) Quantitative analysis of signal co-localization between mitochondrial nucleoids and early endosomes in patient and control fibroblasts determined by Pearson’s coefficient, lines indicate mean +/- SD, points indicate replicates. (D) Pathway model representing mitochondrial dysfunction leading to sterile inflammation. The diagram shows how the MFN2 Q367H variant can alter mitochondrial function, leading to transfer of mtDNA from the mitochondria to the early endosomes (middle). This can lead to detection by TLR9 and subsequent activation of the NF-kB mediated inflammation (right). Additionally, mtDNA in the early endosomes could leak into the cytosol, leading to detection through the cGAS-STING pathway (middle-bottom), promoting Type-I Interferon mediated inflammation (right). Collectively, the inflammation contributes to the distal myopathy disease outcome in the patient. P-values determined by unpaired t-test (ns=not significant (p>0.05); *=p≤ 0.05; \**\*|\*=p≤ 0.0001).

Given the link to TLR9 signalling, which detects mtDNA in endosomes, we also looked to see if mtDNA outside the mitochondrial network colocalized with Rab5C-labelled early endosomes. Remarkably, we observed co-localization of extra-mitochondrial mtDNA with early endosomes, including enlarged endosomes harbouring several mtDNA nucleoids (Figure 6B-C). Overall, these findings point to a mtDNA-mediated sterile inflammatory response, mediated through localization of mtDNA into early endosomes and the cytosol (Figure 6D).

### MFN2 Q367H Patient Transdifferentiated Myoblasts Show Elevated Inflammation, Disrupted MERCs and Reduced mtDNA Nucleoid Content

To examine the cellular and inflammatory phenotypes more closely in a relevant cell type, we transdifferentiated control and patient fibroblasts into myoblasts for additional characterization. Upon transdifferentiating, we still observed increased inflammatory signalling (Figure 7A). However, compared to fibroblast cells, patient myoblasts exhibited an additional 3-fold increase in cGAS-STING interferons, and a 10-fold increase in the TLR9 inflammatory signalling, indicating that these pathways are even more strongly activated in patient muscle cells. Finally, we examined some basic parameters of MFN2 functions in our transdifferentiated patient myoblasts. First, we found no changes in gross mitochondrial morphology or the average length of mitochondrial branches in patient myoblasts (Figure 7B-C). Second, patient myoblasts had fewer MERCs, which were also smaller in size (Figure 7D-F). Finally, the mtDNA nucleoid analysis found a reduction in the number nucleoids per cell in patient myoblasts, as well as a decrease in size of the nucleoids (Figure 7G-I). Notably, consistent with findings in patient fibroblasts, a subset of these nucleoids could be found outside of the mitochondrial network.

![Figure 7:](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2024/06/21/2024.06.20.24309123/F7.medium.gif)

[Figure 7:](http://medrxiv.org/content/early/2024/06/21/2024.06.20.24309123/F7)

Figure 7: Altered MFN2 functions in transdifferentiated patient myoblasts.
(A) RT-qPCR data showing gene expression for TLR9/NF-KB targets (Asc, IL6, NLRP3, S100a9, TNF) and cGAS-STING targets (Ifn-α, Ifn-β) in control vs patient myoblasts, lines indicate mean +/- SD.(B) Representative confocal images showing mitochondrial network morphology (TOMM20; cyan) in control and patient myoblasts. (C) Quantitative analysis of mean mitochondrial branch length in control and patient myoblasts, violin plots show median and interquartile ranges. (D) Representative confocal images of control and patient myoblasts showing PLA probes (representing MERCs; green), mitochondria (TOMM20; cyan) and ER (Calnexin; magenta). (E) Quantitative analysis of the number of MERCs per cell in control and patient myoblasts, violin plots show media and interquartile ranges. (F) Quantitative analysis of average size of MERCs per cell in control and patient myoblasts, violin plots show media and interquartile ranges. (G) Representative confocal images showing mitochondria (TOMM20) and mitochondrial nucleoids (dsDNA antibody). Circles indicate mtDNA staining outside the mitochondrial network. (H) Quantification of average number of nucleoids per cell in control and patient myoblasts, violin plots show median and interquartile ranges. (I) Quantification of average size of mtDNA nucleoids in control and patient myoblasts. Violin plots show median and interquartile ranges. P-values determined by unpaired t-test (ns=not significant (p>0.05); \***|=p≤ 0.001; \**\*|\*=p≤ 0.0001).

## Discussion

We report an individual with a late-onset myopathy phenotype and a previously uncharacterized variant in *MFN2*. We pursued the investigations described in this report based on strong clinical suspicion for this as a potential pathogenic variant. This sequence change in *MFN2* replaces glutamine with histidine at codon 367 of the MFN2 protein (Q367H). The glutamine residue is highly conserved, and there is a small physicochemical difference between glutamine and histidine, implying a hypermorphic change. The MFN2 Q367 amino acid is conserved across vertebrate MFN2 sequences, as well as with the MFN1 paralogue (Figure 1D). The MFN2 Q367H variant is present in population databases with a very low allele frequency (highest allele frequency 0.000242 in Gnomad according to dbSNP report rs373211062, accessed June 4, 2024), including one homozygous individual. The variant has an entry in Clinvar (439898, accessed June 4, 2024) and is classified as having conflicting interpretations of pathogenicity, with 4 entries from clinical testing of patients with neurological phenotypes. In a large cohort study, a patient with early onset peripheral neuropathy was reported who was compound heterozygous for Q367H and P184R MFN2 variants (42). However, the Q367H MFN2 variant has not been functionally characterized, or previously reported in the literature as a standalone disease-causing variant. The family history of the patient’s parent suggests autosomal dominant inheritance, which is in keeping with this variant, although confirmation of this variant in the parent was not possible. Furthermore, a different amino acid substitution at this position (Q367P) was associated with a neuropathy presentation (43), providing additional support that variants at this position are likely to be associated with this disease. All considered, the Q367H variant is highly plausible as a reduced penetrance pathogenic variant given its extremely low frequency in population databases, its prevalence in individuals with neurological phenotypes, its presence in published cases from literature. The data from studies in this report point to the deleterious nature of the Q367H variant, and importantly demonstrate the damaging effect of this variant in cells transdifferentiated to the muscle lineage.

This case illustrates many of the challenges when conducting research with people affected by genetic conditions with very late onset. Information from control population databases may be misleading if individuals are presymptomatic for a very late onset disease, or when phenotypes are very mild and slowly progressive. Even for paediatric-onset disease, pathogenic variants appear in a substantial proportion of control individuals (2.8% in one study)(44), and this number may be comparable or even higher for very-late onset disorders, such as the one described in this report. Future research will improve our understanding of genetic variants and disorders with very late onset and variable penetrance, but identification and investigation of cases is difficult since they may be mistaken for degenerative or acquired conditions (for example, prior work showing diagnoses of genetic spastic paraplegia in patients initially diagnosed with sporadic primary lateral sclerosis)(45). Other factors include the lack of availability of family members for segregation of variants (i.e: parents have expired by the time the patient is diagnosed, or family members are estranged, or the disease status is not yet known for younger family members). In this respect, we acknowledge limitations of this study which only includes a single affected participant. However, we hope to bring attention to the fact that the investigation of genetic disorders is important to our broader understanding of genetic diseases and operates with the above-mentioned special challenges.

Despite the fact that over 150 MFN2 variants have been reported in patients with CMT2A, we still do not have a complete understanding of how MFN2 dysfunction contributes to disease. The rfsituation is complex, as MFN2 has multiple functions in the cell, and CMT2A can have additional pathological phenotypes beyond peripheral neuropathy. Moreover, only a handful of MFN2 variants have been characterized functionally, and even then, often only for one or two of MFN2’s known functions (16). Thus, a better understanding of how MFN2 variants affect MFN2 functions is essential to understanding how MFN2 contributes to disease. Here, our characterization of cells from a patient exhibiting distal myopathy who carried the candidate pathogenic MFN2 Q367H variant provides novel mechanistic insight into MFN2-mediated pathology. The fact that several MFN2-mediated functions are impaired in the patient cells, especially when they are forced to use mitochondria for energy production, supports the notion that the Q367H variant impairs MFN2 function and likely contributes to the patient pathology. Moreover, to our knowledge, the association of mtDNA release and inflammation with an MFN2 variant has not been reported previously, and supports a mechanistic pathway by which MFN2 dysfunction contributes to inflammation. Finally, the elevated levels of inflammation in patient-derived transdifferentiated myoblasts further support a novel disease pathomechanism for MFN2-mediated inflammation driving myopathy.

Several lines of evidence support the pathogenicity of the Q367H MFN2 variant, as patient fibroblasts show perturbation to multiple functions performed by MFN2, including mitochondrial fusion mtDNA nucleoid distribution, MERCs and cellular lipid droplets. With respect to mitochondrial fusion, similar to our findings, several MFN2 variants do not show mitochondrial fragmentation when grown in standard glucose conditions (16). However, culturing fibroblasts in galactose can reveal fragmentation of the mitochondrial network, as reported for MFN2 D210V (9), and as we see for the Q367H variant. This network fragmentation could be caused by reduced fusion and/or increased fission (46). Given the well-defined role of MFN2 in mitochondrial fusion, the fragmentation is consistent with reduced fusion. Meanwhile, as reduced mitochondrial fusion can also impact mtDNA distribution (37), the changes we see in mitochondrial nucleoids are also consistent with reduced fusion in patient fibroblasts when grown in galactose. Taken together, the results are consistent with reduced mitochondrial fusion when cells are grown in galactose.

Beyond mitochondrial fusion, perhaps the best characterized function of MFN2 is in regulating MERCs (2). Notably, MERCs in fibroblasts from patients with various *MFN2* variants can show a variety of changes (16), and the reduction in the number and size of MERCs that we observed in patient cells is consistent with MFN2 dysfunction. Intriguingly, altered MERCs could contribute to the patient myopathy, as pathogenic variants in *SEPN1*, another gene implicated in MERC biology, also causes a myopathy phenotype (47).

Also consistent with MFN2 dysfunction and in line with our findings, an increased number and size of lipid droplets in fibroblasts is reported in fibroblasts patients with other MFN2 variants (48), as well as upon MFN2 depletion (3). Intriguingly, the lipid droplet phenotype is evident in standard growth conditions when mitochondrial morphology was unaltered. Notably, the mechanism by which impaired MFN2 function leads to increased size and abundance of lipid droplets in patient fibroblasts has not been investigated previously. We envision several possible explanations: decreased transfer of lipids into mitochondria from lipid droplets, or increased transfer of lipids from mitochondria to lipid droplets. Our finding that patient fibroblasts failed to import fatty acids into mitochondria, even under glucose starvation conditions, is in line with the first possibility, and likely reflects the role of MFN2 in mediating mitochondria-lipid droplet interactions (49). Though we expect reduced lipid transfer will also explain the increased lipid droplet accumulation observed in fibroblasts from patients with other MFN2 variants, this remains to be determined.

Characterization of the transdifferentiated patient myoblasts provides further evidence of MFN2 dysfunction and important insight into the muscle myopathy phenotype in the patient. Consistent with MFN2 dysfunction, we see a significant decrease in the number and size of mitochondrial nucleoids, as well as fewer and smaller MERCs in transdifferentiated patient myoblasts. However, we did not observe any changes in mitochondrial network morphology. Notably, the transdifferentiated myoblasts were grown in standard myoblast media, as opposed to the glucose-free galactose-supplemented media required to elicit differences in the patient fibroblasts. Other differences in growth conditions between transdifferentiated patient myoblasts (*e.g.*, collagen matrix), or the inherent differences between fibroblast and myoblast cell types may also explain differences between patient fibroblasts and transdifferentiated myoblasts. Regardless, the reduced number of nucleoids in the patient cells is consistent with MFN2 dysfunction, as muscle-specific loss of MFN2 leads to mtDNA depletion (17).

The finding of mtDNA release and inflammation in both Q367H patient fibroblasts and transdifferentiated myoblasts is novel in the context of pathogenic MFN2 variants and is likely relevant to disease mechanism. The fact that inhibitors of both TLR9/cGAS-STING pathways lead to decrease in their downstream cytokines, indicates that both pathways are activated. These findings are also consistent with previous work on impaired mitochondrial fusion and increased inflammation. With respect to MFNs, a recent paper by Irazoki *et al.*, found that knockdown of either MFN1 or MFN2 in myoblasts could promote TLR9 signalling (4). Arguing that activation of TLR9-mediated inflammation is a general response to loss of mitochondrial fusion, loss of the inner mitochondrial membrane fusion protein OPA1 in muscle leads to TLR9-mediated NF-κB activation (but not cGAS-STING), where it also drives an inflammatory myopathy (50). Thus, it seems that loss of fusion is sufficient to induce mtDNA-mediated TLR9 inflammation in muscle cells that can drive myopathy. A mechanism that can explain the phenotype of the patient harbouring the Q367H variant. Meanwhile, the increased cGAS-STING signalling that we also see in Q367H patient cells is consistent with increased cGAS-STING activity also reported in MFN2-deficient myoblasts (4), as well as in microglia where MFN2 is downregulated (51). Whether cGAS-STING signalling further perturbs pathology remains to be determined, though elevated cGAS-STING inflammation is also linked to myopathy (52).

Another interesting question involves how MFN2 loss of function drives mtDNA release to stimulate both TLR9 and cGAS-STING inflammation. In this regard, several recent studies are relevant to endosomal-mediated mtDNA release (53–55). A model seems to be emerging where, in the context of stalled mtDNA replication intermediates, enlarged nucleoid clusters are transferred directly into early endosomes, where the mtDNA can be detected by TLR9. When this quality control pathway is overwhelmed, it can lead to release of mtDNA from endosomes into the cytosol, and cGAS-STING activation (54). Notably, one mechanism by which loss of mitochondrial fusion is proposed to impact the mtDNA is through imbalanced distribution of mtDNA maintenance proteins (56), which may lead to an increased abundance of stalled mtDNA replication intermediates. Thus, mtDNA replication fork stalling in response to impaired mitochondrial fusion may trigger initial mtDNA release into endosomes, and the subsequent TLR9 activation associated with MFN2 dysfunction. The fact that we see extra mtDNA and inflammatory signalling in patient cells with normal mitochondrial morphology (*e.g*., fibroblasts grown in standard media, and differentiated myoblasts), suggests that even subtle alterations in mitochondrial fusion are sufficient to impact the mtDNA. Alternatively (or in conjunction), the differences in MERCs, which can demarcate sites of mtDNA replication (57,58), could also impair mtDNA replication and trigger to mtDNA release into endosomes.

Meanwhile, as elevated cGAS-STING is not found with MFN1 or OPA1 depletion (4,50,59), the elevated cGAS-STING signalling in myoblasts devoid of MFN2 (4), and seen here in MFN2 Q367H patient cells, suggests another role for MFN2 in mediating mtDNA release beyond just reduced mitochondrial fusion. In this regard, MFN2 can interact with RAB5C on early endosomes (4). As such, it is tempting to speculate that MFN2 has a role in mediating the mitochondrial-endosome interactions required for endosome-mediated removal of stalled mtDNA replication intermediates, and that disrupting this interaction promotes mtDNA escape into the cytosol to activate cGAS-STING inflammation. Future studies examining the mechanisms of mtDNA release will likely provide more insight to the role played by MFN2 and may benefit from examining the Q367H variant as a means to perturb MFN2 function.

It is worth noting that MFN2 could also influence inflammation through its other functions. For example, there are connections between disrupted MERCs and activation of sterile inflammation, particularly in high energy-demanding tissues such as the heart (60). Meanwhile, relevant to the increased lipid droplet abundance associated with MFN2 dysfunction and seen even in standard growth conditions with normal mitochondrial morphology, it is notable that lipid droplet biology intersects with inflammation. Specifically, elevated lipid droplet abundance correlates with increased inflammation (61–64). Overall, our findings are consistent with MFN2 dysfunction contributing to inflammation, which is a possible explanation for myopathy initially described in the proband.

The patient presentation, with late onset myopathy, but without neuropathy, offers a unique opportunity to separate the pathological outcomes of MFN2 dysfunction by examining the specific mechanism leading to myopathy. Although myopathy has been reported previously in patients with CMT2A (10,11), it has always been in the context of a peripheral neuropathy where it is generally considered to be a downstream consequence of nerve loss. For example, the MFN2 D210V variant has myopathy phenotypes in a three-generational case study, which has accompanying axonal neuropathy (9). On the other hand, arguing against the notion of muscle loss downstream of neuropathy, mitochondrial myopathy is seen in a skeletal muscle-specific MFN2 loss-of-function study in animal models, highlighting the importance of the protein in tissues such as skeletal muscle (65,66). Meanwhile, in cardiac muscle, the MFN2 R400Q variant is implicated in cardiomyopathy, though no link to inflammation was investigated (67). Additionally, the loss of MFN2 protein expression in skeletal muscles correlates with age and could contribute to sarcopenia through a similar inflammatory mechanism (68). Finally, similar to the patient we describe here, a novel ENU-induced Mfn2 mouse model characterized L643P as a variant that causes muscle myopathy without peripheral neuropathy (69). Together, these findings show that MFN2 dysfunction can cause muscle myopathy in the absence of nerve degeneration, as observed in the proband.

With respect to how MFN2 dysfunction might cause myopathy, in the first report of an MFN2-specific mouse muscle knockout, the authors argued that the mtDNA depletion was likely a key contributory factor due to reduced oxidative phosphorylation (65). However, more recent work linking MFN1 and OPA1 muscle deficiency to inflammation mediated myopathy, showed that blocking the inflammation was protective in these models (4,50), arguing for a key role of inflammation in the muscle pathology caused by impaired mitochondrial fusion. Mechanistically, inflammation in the muscle cell niche can lead to myopathy and interfere with muscle stem cell populations (70), as well as affect the expression of myogenic genes (71). Meanwhile, NF-kB, a proinflammatory cytokine that is downstream of TLR9 signaling, can cause muscle wasting in mice (72). Building on these findings, it is tempting to speculate that mtDNA-mediated inflammation may also be relevant to idiopathic inflammatory myopathies, a heterogeneous group of autoimmune disorders with varying clinical manifestations, primarily defined by muscle weakness (73,74). In this regard, we speculate that the mtDNA-mediated inflammation discovered in Q367H patient-derived cells is contributing to the patient’s myopathy.

Although MFN1 and OPA1 deficiency can cause inflammatory myopathy in mouse muscle knockout models, the role of mtDNA-mediated inflammation in MFN2 pathology has not been fully explored. While a mouse knock-in model of the MFN2 K357T variant is linked to augmented neuroinflammation in response to LPS treatment, the role of mtDNA was not investigated (75). Our findings linking the MFN2 Q367H variant to mtDNA release and inflammation thus provide a novel disease mechanism for MFN2 dysfunction. Moreover, while our findings are relevant to the myopathy phenotype in muscle, it is also possible that mtDNA-mediated inflammation contributes to the peripheral neuropathy in neurons. Intriguingly, peripheral neuropathy is a common phenotype of several DNA damaging agents (76–80), such as doxorubicin, which can damage mtDNA and lead to mtDNA-mediated cGAS-STING inflammation (81).

In summary, we observed elevated mtDNA-mediated TLR9/NF-kB and cGAS-STING inflammatory signaling in a patient harboring the uncharacterized MFN2 Q367H variant who displayed late-onset myopathy as a standalone phenotype. Moreover, functional analyses of patient fibroblasts provided several lines of evidence consistent with MFN2 dysfunction. Critically, characterization of transdifferentiated myoblasts and the elucidation of a novel link to mtDNA-mediated inflammation may mechanistically explain the myopathy phenotype in the patient.

## Supporting information

Supplemental Figure 1 [[supplements/309123_file11.pdf]](pending:yes)

## Data Availability

All data produced in the present work are contained in the manuscript

## Author contributions

Designed and/or performed and analyzed mitochondrial experiments, MZ, GS, WA, TS, MJ, RS, GP, and TES; Clinical evaluation of patients, GP; Analyzed exome data GP, MJ,; Performed skin biopsy from mitochondrial patients, GP; Drafted the manuscript or figures, MZ, GP, TES. All authors discussed the results and commented on the manuscript. GS, and WA are joint second authors. TES and GP are co-corresponding authors.

## Funding

This work was supported by funds provided by the Alberta Children’s Hospital Research Institute (TES), the Canadian Institutes of Health Research (TES), and the New Frontiers Research Fund (TES). MZ was supported by a Hotchkiss Brain Institute International Recruitment Scholarship. WA was supported by the Saudi Cultural Bureau. R.S. was supported by a QEII Graduate Scholarship. TGBS was supported by an Alberta Graduate Excellence Scholarship and a University of Calgary Faculty of Graduate Studies Doctoral Scholarship. MJ is the recipient of the Katharine Sarah Melinda Mei-Ling Thomas Rare Diseases Scholarship. Infrastructure in the lab of GP for this work was supported by the HBI Dementia Research Equipment Fund and a Canada Foundation for Innovation JELF Grant (Neuromuscular Genetics, 36624). The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

## Acknowledgements

The authors would like to thank the study participants and their family. Lentivirus production was performed by the Hotchkiss Brain Institute Molecular Core Facility at the University of Calgary.

*   Received June 20, 2024.
*   Revision received June 20, 2024.
*   Accepted June 21, 2024.


*   © 2024, Posted by Cold Spring Harbor Laboratory

This pre-print is available under a Creative Commons License (Attribution-NonCommercial 4.0 International), CC BY-NC 4.0, as described at [http://creativecommons.org/licenses/by-nc/4.0/](http://creativecommons.org/licenses/by-nc/4.0/)

## References

1.  1.Filadi R, Pendin D, Pizzo P. Mitofusin 2: from functions to disease. Cell Death Dis. 2018 Feb 28;9(3):330.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41419-017-0023-6&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=29491355&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 

2.  2.de Brito OM, Scorrano L. Mitofusin 2 tethers endoplasmic reticulum to mitochondria. Nature. 2008 Dec 4;456(7222):605–10.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/nature07534&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=19052620&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000261340000033&link_type=ISI) 

3.  3.Boutant M, Kulkarni SS, Joffraud M, Ratajczak J, Valera-Alberni M, Combe R, et al. Mfn2 is critical for brown adipose tissue thermogenic function. EMBO J. 2017 Jun 1;36(11):1543–58.
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZW1ib2pubCI7czo1OiJyZXNpZCI7czoxMDoiMzYvMTEvMTU0MyI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDI0LzA2LzIxLzIwMjQuMDYuMjAuMjQzMDkxMjMuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

4.  4.Irazoki A, Gordaliza-Alaguero I, Frank E, Giakoumakis NN, Seco J, Palacín M, et al. Disruption of mitochondrial dynamics triggers muscle inflammation through interorganellar contacts and mitochondrial DNA mislocation. Nat Commun. 2023 Jan 6;14(1):108.
    
    

5.  5.Zervopoulos SD, Boukouris AE, Saleme B, Haromy A, Tejay S, Sutendra G, et al. MFN2-driven mitochondria-to-nucleus tethering allows a non-canonical nuclear entry pathway of the mitochondrial pyruvate dehydrogenase complex. Mol Cell. 2022 Mar 3;82(5):1066–1077.e7.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.molcel.2022.02.003&link_type=DOI) 

6.  6.Li J, Dang X, Franco A, Dorn GW. Reciprocal Regulation of Mitofusin 2-Mediated Mitophagy and Mitochondrial Fusion by Different PINK1 Phosphorylation Events. Front Cell Dev Biol. 2022;10:868465.
    
    

7.  7.McLelland GL, Goiran T, Yi W, Dorval G, Chen CX, Lauinger ND, et al. Mfn2 ubiquitination by PINK1/parkin gates the p97-dependent release of ER from mitochondria to drive mitophagy. Dikic I, editor. eLife. 2018 Apr 20;7:e32866.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.7554/elife.32866&link_type=DOI) 

8.  8.Feely SME, Laura M, Siskind CE, Sottile S, Davis M, Gibbons VS, et al. MFN2 mutations cause severe phenotypes in most patients with CMT2A. Neurology. 2011 May 17;76(20):1690–6.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1212/WNL.0b013e31821a441e&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21508331&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 

9.  9.Rouzier C, Bannwarth S, Chaussenot A, Chevrollier A, Verschueren A, Bonello-Palot N, et al. The MFN2 gene is responsible for mitochondrial DNA instability and optic atrophy “plus” phenotype. Brain J Neurol. 2012 Jan;135(Pt 1):23–34.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/brain/awr323&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=22189565&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000300044400012&link_type=ISI) 

10. 10.Abati E, Manini A, Velardo D, Del Bo R, Napoli L, Rizzo F, et al. Clinical and genetic features of a cohort of patients with MFN2-related neuropathy. Sci Rep. 2022 Apr 13;12(1):6181.
    
    

11. 11.Verhoeven K, Claeys KG, Züchner S, Schröder JM, Weis J, Ceuterick C, et al. MFN2 mutation distribution and genotype/phenotype correlation in Charcot–Marie–Tooth type 2. Brain. 2006 Aug 1;129(8):2093–102.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/brain/awl126&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=16714318&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000239900600018&link_type=ISI) 

12. 12.Sharma G, Zaman M, Sabouny R, Joel M, Martens K, Martino D, et al. Characterization of a novel variant in the HR1 domain of MFN2 in a patient with ataxia, optic atrophy and sensorineural hearing loss. F1000Research. 2022 Sep 2;10:606.
    
    

13. 13.Sawyer SL, Cheuk-Him Ng A, Innes AM, Wagner JD, Dyment DA, Tetreault M, et al. Homozygous mutations in MFN2 cause multiple symmetric lipomatosis associated with neuropathy. Hum Mol Genet. 2015 Sep 15;24(18):5109–14.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/hmg/ddv229&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=26085578&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 

14. 14.Mann JP, Duan X, Patel S, Tábara LC, Scurria F, Alvarez-Guaita A, et al. A mouse model of human mitofusin-2-related lipodystrophy exhibits adipose-specific mitochondrial stress and reduced leptin secretion. eLife. 2023 Feb 1;12:e82283.
    
    

15. 15.Masingue M, Vatier C, Jéru I, Latour P, Jardel C, Laforêt P, et al. Homozygous p.R707W MFN2 mutation is associated with neuropathy, lipomatosis, peripheral lipoatrophy and metabolic alterations. Neuromuscul Disord. 2017 Oct 1;27:S148.
    
    

16. 16.Zaman M, Shutt TE. The Role of Impaired Mitochondrial Dynamics in MFN2-Mediated Pathology. Front Cell Dev Biol. 2022;10:858286.
    
    

17. 17.Chen H, Vermulst M, Wang YE, Chomyn A, Prolla TA, McCaffery JM, et al. Mitochondrial fusion is required for mtDNA stability in skeletal muscle and tolerance of mtDNA mutations. Cell. 2010 Apr 16;141(2):280–9.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.cell.2010.02.026&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=20403324&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000276738400018&link_type=ISI) 

18. 18.Rodríguez-Nuevo A, Díaz-Ramos A, Noguera E, Díaz-Sáez F, Duran X, Muñoz JP, et al. Mitochondrial DNA and TLR9 drive muscle inflammation upon Opa1 deficiency. EMBO J. 2018 May 15;37(10):e96553.
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZW1ib2pubCI7czo1OiJyZXNpZCI7czoxMjoiMzcvMTAvZTk2NTUzIjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjQvMDYvMjEvMjAyNC4wNi4yMC4yNDMwOTEyMy5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 

19. 19.Sidarala V, Zhu J, Levi-D’Ancona E, Pearson GL, Reck EC, Walker EM, et al. Mitofusin 1 and 2 regulation of mitochondrial DNA content is a critical determinant of glucose homeostasis. Nat Commun. 2022 Apr 29;13(1):2340.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41467-022-29945-7&link_type=DOI) 

20. 20.Vielhaber S, Debska-Vielhaber G, Peeva V, Schoeler S, Kudin AP, Minin I, et al. Mitofusin 2 mutations affect mitochondrial function by mitochondrial DNA depletion. Acta Neuropathol (Berl). 2013 Feb;125(2):245–56.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s00401-012-1036-y&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=22926664&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000313793400006&link_type=ISI) 

21. 21.Al Khatib I, Deng J, Lei Y, Torres-Odio S, Rojas GR, Newman LE, et al. Activation of the cGAS-STING innate immune response in cells with deficient mitochondrial topoisomerase TOP1MT. Hum Mol Genet. 2023 Jul 20;32(15):2422–40.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/hmg/ddad062&link_type=DOI) 

22. 22.Kim J, Kim HS, Chung JH. Molecular mechanisms of mitochondrial DNA release and activation of the cGAS-STING pathway. Exp Mol Med. 2023 Mar;55(3):510–9.
    
    

23. 23.Pereira CA, Carlos D, Ferreira NS, Silva JF, Zanotto CZ, Zamboni DS, et al. Mitochondrial DNA Promotes NLRP3 Inflammasome Activation and Contributes to Endothelial Dysfunction and Inflammation in Type 1 Diabetes. Front Physiol. 2019;10:1557.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3389/fphys.2019.01557&link_type=DOI) 

24. 24.Qiu Y, Huang Y, Chen M, Yang Y, Li X, Zhang W. Mitochondrial DNA in NLRP3 inflammasome activation. Int Immunopharmacol. 2022 Jul;108:108719.
    
    

25. 25.Martens K, Leckie J, Fok D, Wells RA, Chhibber S, Pfeffer G. Case Report: Calpainopathy Presenting After Bone Marrow Transplantation, With Studies of Donor Genetic Content in Various Tissue Types. Front Neurol. 2020;11:604547.
    
    

26. 26.Griffin HR, Pyle A, Blakely EL, Alston CL, Duff J, Hudson G, et al. Accurate mitochondrial DNA sequencing using off-target reads provides a single test to identify pathogenic point mutations. Genet Med. 2014 Dec;16(12):962–71.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/gim.2014.66&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=24901348&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 

27. 27.Di Tommaso P, Moretti S, Xenarios I, Orobitg M, Montanyola A, Chang JM, et al. T-Coffee: a web server for the multiple sequence alignment of protein and RNA sequences using structural information and homology extension. Nucleic Acids Res. 2011 Jul;39(Web Server issue):W13–17.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/nar/gkr245&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21558174&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000292325300003&link_type=ISI) 

28. 28.UniProt Consortium. UniProt: the Universal Protein Knowledgebase in 2023. Nucleic Acids Res. 2023 Jan 6;51(D1):D523–31.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/NAR/GKAC1052&link_type=DOI) 

29. 29.Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, et al. Fiji: an open-source platform for biological-image analysis. Nat Methods. 2012 Jun 28;9(7):676–82.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/nmeth.2019&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=22743772&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000305942200021&link_type=ISI) 

30. 30.Uddin GM, Lacroix R, Zaman M, Khatib IA, Rho JM, Kurrasch DM, et al. The Ketogenic Diet Metabolite β-Hydroxybutyrate Promotes Mitochondrial Elongation via Deacetylation and Improves Autism-like Behavior in Zebrafish. bioRxiv. 2022 Jan 1;2022.10.03.510695.
    
    

31. 31.Zhao T, Goedhart CM, Sam PN, Sabouny R, Lingrell S, Cornish AJ, et al. PISD is a mitochondrial disease gene causing skeletal dysplasia, cataracts, and white matter changes. Life Sci Alliance. 2019 Apr;2(2):e201900353.
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6MzoibHNhIjtzOjU6InJlc2lkIjtzOjE0OiIyLzIvZTIwMTkwMDM1MyI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDI0LzA2LzIxLzIwMjQuMDYuMjAuMjQzMDkxMjMuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

32. 32.Al Khatib I, Deng J, Symes A, Kerr M, Zhang H, Huang SYN, et al. Functional characterization of two variants of mitochondrial topoisomerase TOP1MT that impact regulation of the mitochondrial genome. J Biol Chem. 2022 Oct;298(10):102420.
    
    

33. 33.Valente AJ, Maddalena LA, Robb EL, Moradi F, Stuart JA. A simple ImageJ macro tool for analyzing mitochondrial network morphology in mammalian cell culture. Acta Histochem. 2017 Apr;119(3):315–26.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.acthis.2017.03.001&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=28314612&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 

34. 34.Bolte S, Cordelières FP. A guided tour into subcellular colocalization analysis in light microscopy. J Microsc. 2006 Dec;224(Pt 3):213–32.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/j.1365-2818.2006.01706.x&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=17210054&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000242993500001&link_type=ISI) 

35. 35.Zhang Q, Bethmann C, Worth NF, Davies JD, Wasner C, Feuer A, et al. Nesprin-1 and -2 are involved in the pathogenesis of Emery Dreifuss muscular dystrophy and are critical for nuclear envelope integrity. Hum Mol Genet. 2007 Dec 1;16(23):2816–33.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/hmg/ddm238&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=17761684&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000251036400004&link_type=ISI) 

36. 36.Beręsewicz M, Boratyńska-Jasińska A, Charzewski Ł, Kawalec M, Kabzińska D, Kochański A, et al. The Effect of a Novel c.820C>T (Arg274Trp) Mutation in the Mitofusin 2 Gene on Fibroblast Metabolism and Clinical Manifestation in a Patient. PloS One. 2017;12(1):e0169999.
    
    

37. 37.Sabouny R, Shutt TE. The role of mitochondrial dynamics in mtDNA maintenance. J Cell Sci. 2021 Dec 15;134(24):jcs258944.
    
    

38. 38.de Brito OM, Scorrano L. Mitofusin 2 tethers endoplasmic reticulum to mitochondria. Nature. 2008 Dec 4;456(7222):605–10.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/nature07534&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=19052620&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000261340000033&link_type=ISI) 

39. 39.Rambold AS, Cohen S, Lippincott-Schwartz J. Fatty acid trafficking in starved cells: regulation by lipid droplet lipolysis, autophagy, and mitochondrial fusion dynamics. Dev Cell. 2015 Mar 23;32(6):678–92.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.devcel.2015.01.029&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=25752962&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 

40. 40.Yasuda H, Leelahavanichkul A, Tsunoda S, Dear JW, Takahashi Y, Ito S, et al. Chloroquine and inhibition of Toll-like receptor 9 protect from sepsis-induced acute kidney injury. Am J Physiol Renal Physiol. 2008 May;294(5):F1050–8.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1152/ajprenal.00461.2007&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=18305095&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000255501200004&link_type=ISI) 

41. 41.Vincent J, Adura C, Gao P, Luz A, Lama L, Asano Y, et al. Small molecule inhibition of cGAS reduces interferon expression in primary macrophages from autoimmune mice. Nat Commun. 2017 Sep 29;8(1):750.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41467-017-00833-9&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=28963528&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 

42. 42.Pipis M, Feely SME, Polke JM, Skorupinska M, Perez L, Shy RR, et al. Natural history of Charcot-Marie-Tooth disease type 2A: a large international multicentre study. Brain J Neurol. 2020 Dec 1;143(12):3589–602.
    
    

43. 43.Xie Y, Li X, Liu L, Hu Z, Huang S, Zhan Y, et al. MFN2-related genetic and clinical features in a cohort of Chinese CMT2 patients. J Peripher Nerv Syst JPNS. 2016 Mar;21(1):38–44.
    
    

44. 44.Tarailo-Graovac M, Zhu JYA, Matthews A, van Karnebeek CDM, Wasserman WW. Assessment of the ExAC data set for the presence of individuals with pathogenic genotypes implicated in severe Mendelian pediatric disorders. Genet Med Off J Am Coll Med Genet. 2017 Dec;19(12):1300–8.
    
    

45. 45.Almomen M, Martens K, Quadir A, Pontifex CS, Hanson A, Korngut L, et al. High diagnostic yield and novel variants in very late-onset spasticity. J Neurogenet. 2019 Mar;33(1):27–32.
    
    

46. 46.Sharma G, Pfeffer G, Shutt TE. Genetic Neuropathy Due to Impairments in Mitochondrial Dynamics. Biology. 2021 Apr;10(4):268.
    
    

47. 47.Filipe A, Chernorudskiy A, Arbogast S, Varone E, Villar-Quiles RN, Pozzer D, et al. Defective endoplasmic reticulum-mitochondria contacts and bioenergetics in SEPN1-related myopathy. Cell Death Differ. 2021 Jan;28(1):123.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41418-020-0587-z&link_type=DOI) 

48. 48.Larrea D, Pera M, Gonnelli A, Quintana-Cabrera R, Akman HO, Guardia-Laguarta C, et al. MFN2 mutations in Charcot-Marie-Tooth disease alter mitochondria-associated ER membrane function but do not impair bioenergetics. Hum Mol Genet. 2019 Jun 1;28(11):1782–800.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/hmg/ddz008&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=30649465&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 

49. 49.Boutant M, Kulkarni SS, Joffraud M, Ratajczak J, Valera-Alberni M, Combe R, et al. Mfn2 is critical for brown adipose tissue thermogenic function. EMBO J. 2017 Jun 1;36(11):1543–58.
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZW1ib2pubCI7czo1OiJyZXNpZCI7czoxMDoiMzYvMTEvMTU0MyI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDI0LzA2LzIxLzIwMjQuMDYuMjAuMjQzMDkxMjMuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

50. 50.Rodríguez-Nuevo A, Díaz-Ramos A, Noguera E, Díaz-Sáez F, Duran X, Muñoz JP, et al. Mitochondrial DNA and TLR9 drive muscle inflammation upon Opa1 deficiency. EMBO J. 2018 May 15;37(10):e96553.
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZW1ib2pubCI7czo1OiJyZXNpZCI7czoxMjoiMzcvMTAvZTk2NTUzIjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjQvMDYvMjEvMjAyNC4wNi4yMC4yNDMwOTEyMy5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 

51. 51.Wei FL, Wang TF, Wang CL, Zhang ZP, Zhao JW, Heng W, et al. Cytoplasmic Escape of Mitochondrial DNA Mediated by Mfn2 Downregulation Promotes Microglial Activation via cGas-Sting Axis in Spinal Cord Injury. Adv Sci Weinh Baden-Wurtt Ger. 2023 Nov 27;e2305442.
    
    

52. 52.Zhou M, Cheng X, Zhu W, Jiang J, Zhu S, Wu X, et al. Activation of cGAS-STING pathway - A possible cause of myofiber atrophy/necrosis in dermatomyositis and immune-mediated necrotizing myopathy. J Clin Lab Anal. 2022 Oct;36(10):e24631.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1002/jcla.24631&link_type=DOI) 

53. 53.Sen A, Boix J, Pla-Martín D. Endosomal-dependent mitophagy coordinates mitochondrial nucleoid and mtDNA elimination. Autophagy. 19(9):2609–10.
    
    

54. 54.Newman LE, Tadepalle N, Novak SW, Schiavon CR, Rojas GR, Chevez JA, et al. Endosomal removal and disposal of dysfunctional, immunostimulatory mitochondrial DNA [Internet]. Cell Biology; 2022 Oct [cited 2024 Feb 19]. Available from: [http://biorxiv.org/lookup/doi/10.1101/2022.10.12.511955](http://biorxiv.org/lookup/doi/10.1101/2022.10.12.511955)
    
    

55. 55.Sen A, Kallabis S, Gaedke F, Jüngst C, Boix J, Nüchel J, et al. Mitochondrial membrane proteins and VPS35 orchestrate selective removal of mtDNA. Nat Commun. 2022 Nov 7;13:6704.
    
    

56. 56. Silva Ramos E, Motori E, Brüser C, Kühl I, Yeroslaviz A, Ruzzenente B, et al. Mitochondrial fusion is required for regulation of mitochondrial DNA replication. PLoS Genet. 2019 Jun;15(6):e1008085.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pgen.1008085&link_type=DOI) 

57. 57.Ilamathi HS, Benhammouda S, Lounas A, Al-Naemi K, Desrochers-Goyette J, Lines MA, et al. Contact sites between endoplasmic reticulum sheets and mitochondria regulate mitochondrial DNA replication and segregation. iScience. 2023 Jul 21;26(7):107180.
    
    

58. 58.Lewis SC, Uchiyama LF, Nunnari J. ER-mitochondria contacts couple mtDNA synthesis with mitochondrial division in human cells. Science. 2016 Jul 15;353(6296):aaf5549.
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6Mzoic2NpIjtzOjU6InJlc2lkIjtzOjE2OiIzNTMvNjI5Ni9hYWY1NTQ5IjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjQvMDYvMjEvMjAyNC4wNi4yMC4yNDMwOTEyMy5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 

59. 59.Tur J, Pereira-Lopes S, Vico T, Marín EA, Muñoz JP, Hernández-Alvarez M, et al. Mitofusin 2 in Macrophages Links Mitochondrial ROS Production, Cytokine Release, Phagocytosis, Autophagy, and Bactericidal Activity. Cell Rep. 2020 Aug 25;32(8):108079.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.celrep.2020.108079&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=32846136&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 

60. 60.Liu H, Liu X, Zhuang H, Fan H, Zhu D, Xu Y, et al. Mitochondrial Contact Sites in Inflammation-Induced Cardiovascular Disease. Front Cell Dev Biol. 2020 Jul 30;8:692.
    
    

61. 61.Akhmetova K, Balasov M, Chesnokov I. Drosophila STING protein has a role in lipid metabolism. eLife. 2021 Sep 1;10:e67358.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.7554/eLife.67358&link_type=DOI) 

62. 62.Bai J, Cervantes C, He S, He J, Plasko GR, Wen J, et al. Mitochondrial stress-activated cGAS-STING pathway inhibits thermogenic program and contributes to overnutrition-induced obesity in mice. Commun Biol. 2020 May 22;3(1):257.
    
    

63. 63.Monson EA, Crosse KM, Das M, Helbig KJ. Lipid droplet density alters the early innate immune response to viral infection. Zhang L, editor. PLOS ONE. 2018 Jan 2;13(1):e0190597.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pone.0190597&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=29293661&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 

64. 64.Boucher DM, Vijithakumar V, Ouimet M. Lipid Droplets as Regulators of Metabolism and Immunity. Immunometabolism [Internet]. 2021 Jul [cited 2024 Feb 19];3(3). Available from: [https://journals.lww.com/10.20900/immunometab20210021](https://journals.lww.com/10.20900/immunometab20210021)
    
    

65. 65.Chen H, Vermulst M, Wang YE, Chomyn A, Prolla TA, McCaffery JM, et al. Mitochondrial fusion is required for mtDNA stability in skeletal muscle and tolerance of mtDNA mutations. Cell. 2010 Apr 16;141(2):280–9.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.cell.2010.02.026&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=20403324&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000276738400018&link_type=ISI) 

66. 66.Chen Z, Bordieanu B, Kesavan R, Lesner NP, Venigalla SSK, Shelton SD, et al. Lactate metabolism is essential in early-onset mitochondrial myopathy. Sci Adv. 2023 Jan 4;9(1):eadd3216.
    
    

67. 67.Franco A, Li J, Kelly DP, Hershberger RE, Marian AJ, Lewis RM, et al. A human mitofusin 2 mutation can cause mitophagic cardiomyopathy. eLife. 2023 Nov 1;12:e84235.
    
    

68. 68.Sebastián D, Sorianello E, Segalés J, Irazoki A, Ruiz-Bonilla V, Sala D, et al. Mfn2 deficiency links age-related sarcopenia and impaired autophagy to activation of an adaptive mitophagy pathway. EMBO J. 2016 Aug 1;35(15):1677–93.
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZW1ib2pubCI7czo1OiJyZXNpZCI7czoxMDoiMzUvMTUvMTY3NyI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDI0LzA2LzIxLzIwMjQuMDYuMjAuMjQzMDkxMjMuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

69. 69.Hines TJ, Bailey J, Liu H, Guntur AR, Seburn KL, Pratt SL, et al. A Novel ENU-Induced Mfn2 Mutation Causes Motor Deficits in Mice without Causing Peripheral Neuropathy. Biology. 2023 Jul 3;12(7):953.
    
    

70. 70.Perandini LA, Chimin P, Lutkemeyer D da S, Câmara NOS. Chronic inflammation in skeletal muscle impairs satellite cells function during regeneration: can physical exercise restore the satellite cell niche? FEBS J. 2018 Jun;285(11):1973–84.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/FEBS.14417&link_type=DOI) 

71. 71.Chandran R, Knobloch TJ, Anghelina M, Agarwal S. Biomechanical signals upregulate myogenic gene induction in the presence or absence of inflammation. Am J Physiol Cell Physiol. 2007 Jul;293(1):C267–276.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1152/ajpcell.00594.2006&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=17392379&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000247968900029&link_type=ISI) 

72. 72.Cai D, Frantz JD, Tawa NE, Melendez PA, Oh BC, Lidov HGW, et al. IKKbeta/NF-kappaB activation causes severe muscle wasting in mice. Cell. 2004 Oct 15;119(2):285–98.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.cell.2004.09.027&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=15479644&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000224577200015&link_type=ISI) 

73. 73.Lundberg IE, Fujimoto M, Vencovsky J, Aggarwal R, Holmqvist M, Christopher-Stine L, et al. Idiopathic inflammatory myopathies. Nat Rev Dis Primer. 2021 Dec 2;7(1):86.
    
    

74. 74.Silva AMS, Campos ED, Zanoteli E. Inflammatory myopathies: an update for neurologists. Arq Neuropsiquiatr. 2022 May;80(5 Suppl 1):238–48.
    
    

75. 75.Stavropoulos F, Sargiannidou I, Potamiti L, Kagiava A, Panayiotidis MI, Bae JH, et al. Aberrant Mitochondrial Dynamics and Exacerbated Response to Neuroinflammation in a Novel Mouse Model of CMT2A. Int J Mol Sci. 2021 Oct 26;22(21):11569.
    
    

76. 76.Kerckhove N, Pezet D, Balayssac D. Long-Term Effects, Pathophysiological Mechanisms, and Risk Factors of Chemotherapy-Induced Peripheral Neuropathies: A Comprehensive Literature Review. Front Pharmacol [Internet]. 2017 Feb 24 [cited 2024 Mar 12];8. Available from: [https://www.frontiersin.org/journals/pharmacology/articles/10.3389/fphar.2017.00086/full](https://www.frontiersin.org/journals/pharmacology/articles/10.3389/fphar.2017.00086/full)
    
    

77. 77.Rivera DR, Ganz PA, Weyrich MS, Bandos H, Melnikow J. Chemotherapy-Associated Peripheral Neuropathy in Patients With Early-Stage Breast Cancer: A Systematic Review. JNCI J Natl Cancer Inst. 2018 Feb 1;110(2):djx140.
    
    

78. 78.Jones MR, Urits I, Wolf J, Corrigan D, Colburn L, Peterson E, et al. Drug-Induced Peripheral Neuropathy: A Narrative Review. Curr Clin Pharmacol. 2020 Apr;15(1):38.
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 

79. 79.Grisold W, Cavaletti G, Windebank AJ. Peripheral neuropathies from chemotherapeutics and targeted agents: diagnosis, treatment, and prevention. Neuro-Oncol. 2012 Sep;14(Suppl 4):iv45–54.
    
    

80. 80.Cella D, Huang H, Homesley HD, Montag A, Salani R, De Geest K, et al. Patient-reported peripheral neuropathy of doxorubicin and cisplatin with and without paclitaxel in the treatment of advanced endometrial cancer: Results from GOG 184. Gynecol Oncol. 2010 Dec;119(3):538–42.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.ygyno.2010.08.022&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=20863554&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F06%2F21%2F2024.06.20.24309123.atom) 

81. 81.Wu Z, Oeck S, West AP, Mangalhara KC, Sainz AG, Newman LE, et al. Mitochondrial DNA Stress Signalling Protects the Nuclear Genome. Nat Metab. 2019 Dec;1(12):1209–18.