Abstract
Mutations in nuclear-encoded mitochondrial genes are responsible for a broad spectrum of disorders among which Leigh syndrome (LS) is the most common in infancy. No effective therapies are available for this severe disease mainly because of the limited capabilities of the standard adeno-associated viral (AAV) vectors to transduce both peripheral organs and the central nervous system (CNS) when injected systemically in adults.
Here, we used the brain-penetrating AAV-PHP.B vector to reinstate gene expression in the Ndufs4 KO mouse model of LS. Intravenous delivery of an AAV.PHP.B-Ndufs4 vector in 1-month old KO mice restored mitochondrial complex I activity in several organs including the CNS. This gene replacement strategy extended lifespan, rescued metabolic parameters, provided behavioral improvement, and corrected the pathological phenotype in the brain, retina, and heart of Ndufs4 KO mice. These results provide a robust proof that gene therapy strategies targeting multiple organs can rescue fatal neurometabolic disorders with CNS involvement.
Introduction
Mitochondrial diseases are a heterogenous group of genetic human disorders characterized by a primary defect in the mitochondrial oxidative phosphorylation system. About a third of these diseases results from a deficiency in complex I activity and represents a wide spectrum of clinical manifestations ranging from fatal neonatal diseases to adult-onset neurodegenerative pathologies (Kirby et al., 1999). Leigh syndrome (LS), also known as subacute necrotizing encephalomyelopathy, is the most common infantile mitochondrial disease. This fatal pediatric condition is characterized by gliosis, demyelination and necrotic lesions in multiple brain areas including the brainstem and the basal ganglia that appear as hyperintense signal seen at the MRI. Patients exhibit a variety of symptoms such as psychomotor arrest or decline, ataxia, dystonia, lethargy, abnormal ocular movements and respiratory failure (Kirby et al., 1999). To date, no therapeutic strategy alleviates the pathology and passing often occurs within the decade following diagnosis.
Several mutations have been associated to LS including some in the nuclear gene Ndufs4 that encodes for a non-enzymatic 18kDa subunit of complex I (Vinothkumar et al., 2014). Ndufs4 knockout (KO) mice closely replicate the pathological hallmarks observed in LS patients. While Ndufs4 heterozygous mice appear similar to wild-type animals, KO mice progressively develop encephalopathy leading to an early death around postnatal day (P)50 (Kruse et al., 2008). Starting from P40, mice manifest LS-like symptoms including growth retardation, hypothermia, ataxia, lethargy and breathing irregularities (Kruse et al., 2008). Recent studies also demonstrated the contribution of peripheral organ dysfunction in the phenotypic manifestation highlighting the need for a broad correction of the genetic defect (Quintana et al., 2010; Jin et al., 2014).
At the pre-clinical level, strategies based either on a pharmacological inhibition of mTOR or exposures to hypoxic conditions have provided remarkable therapeutic effects as well as important insights into the disease mechanism but their transposability in the clinic is limited (Johnson et al., 2013; Johnson et al., 2015; Jain et al., 2016; Ferrari et al., 2017). Gene replacement therapy using an AAV9 vector has also been attempted in Ndufs4 KO neonates but resulted in marginal benefits (Quintana et al., 2012; Di Meo et al., 2017).
In the present study we took advantage of the recent development of a novel AAV9 vector variant able to cross the blood brain barrier also in adult mice (Deverman et al., 2016). We therefore used the brain-penetrating AAV-PHP vector for a gene replacement strategy in Ndufs4 KO mice and we showed that this approach afforded a long-term therapeutic effect by correcting LS-associated pathology in several organs and resulted in lifespan extension.
Material and methods
Animals
Ndufs4 heterozygous mice were obtained from Jackson Laboratories (stock number 27058) and bred to produce Ndufs4 KO offsprings. Mice were genotyped at 2 weeks of age using the protocol from Jackson Laboratories and the following primers: wild type forward: AGTCAGCAACATTTTGGCAGT, common: GAGCTTGCCTAGGAGGAGGT, mutant forward: AGGGGACTGGACTAACAGCA. In order to provide warmth and social interaction, KO mice were housed with a minimum of one control littermate. Water bottles had long tip and food was provided on the bottom of each cage containing KO mice so that access to food and water was not a limiting factor for survival. Mice were observed daily and euthanized if they showed a 20% loss in maximum body weight, lethargic behavior, or were found prostate or unconscious. Animals weight was recorded twice a week and body temperature was measured on the same day using a rectal thermometer (Microtherma2, Thermoworks). Measurements were done between 2.00pm and 4.00pm to minimize the effect of circadian variations on body temperature. All experimental procedures were conducted in accordance with the guidelines of the European Directives (2010/63/EU) and the ARRIVE guidelines. They were reviewed and approved by the Italian Ministry of Health, Department of Public Health, Animal Health, Nutrition and Food Safety, in accordance with the law on animal experimentation (article7; D.L. 116/92; protocol number: 26/2014).
Adeno-associated viral (AAV) vector production and injection
The plasmid containing the Ndufs4-IRES-GFP cassette under the control of the CMV-chicken β-actin (CBA) promoter was provided by Pr. Palmiter (University of Washington School of Medicine) (Quintana et al., 2012). The AAV.PHP.B plasmid was provided by Dr. Gradinaru (California Institute of Technology) (Deverman et al., 2016). The AAV-PHP.B-Ndufs4 and the control vector AAV-PHP.B-GFP vector were produced by the AAV Vector Core (Telethon Institute of Genetics and Medicine (TIGEM) as previously described (Colella et al., 2014).
At 4 weeks of age, KO mice and control littermates were randomly assigned to receive an intravenous injection of the Ndufs4-expressing or the GFP control AAV-PHP.B vector (1 × 1012 gc, diluted in phosphate buffer saline (PBS)) through the retro-orbital sinus.
Paw clasping test, cylinder test, hanging test, and seizure frequency
In order to assess neurological decline, the paw clasping test was performed at 45 and 250 days of age. As previously described (Decressac et al., 2010), mice were suspended by the tail and their clasping behavior was examined during 30 seconds.
For the cylinder test, mice were placed in a 2 liters glass cylinder, a video was recorded for 5 min and analyzed using the EthoVision XT software (version 14.0). The number of rears (wall rears, seated rears or free rears; see Fig. 2c and Supplementary videos 1-3) (Schweizer et al., 2016) was scored and the time spend immobile was measured as an index of the locomotor activity.
The hanging test was performed at 45, 100 and 250 days. Mice were placed on the top of a standard wire cage lid. The lid was lightly shaken to cause the animals to grip the wires then turned upside down and suspended at a height of 40 cm over an open cage filled with bedding and excess nesting materials to prevent injury from falling. The latency for the mice to fall off the wire grid was measured and average values were computed from two trials (15 min apart). Trials were stopped if the mouse remained on the lid after 3 min (Luk et al., 2012).
The percentage of animals manifesting epileptic seizure (Supplementary video 4) was estimated at 45, 100 and 250 days. Mice were monitored for the occurrence of epileptic episodes during 3 hours. At the end of an epileptic event, the identity of the mouse was recorded and the animal returned to its home cage.
All testing and analysis were performed by an experimenter blind to the phenotype and treatment.
Electroretinogram (ERG)
The method used for electrophysiological recordings (ERGs) was described previously (Botta et al., 2016). Briefly, mice were dark-adapted for 3 hours. After anaesthesia, mice were accommodated in a stereotaxic apparatus under dim red light, their pupils dilated with a drop of 1% tropicamide (Alcon Laboratories, Inc., Fort Worth, TX) and the body temperature maintained at 37.5°C. ERGs were evoked by flashes of different light intensities ranging from –4 to +1.3 log cd.s/m2, in scotopic condition, generated through a Ganzfeld stimulator (CSO, Florence, Italy). For each animal group, representative a-wave and b-wave track (evoked at + 1.3 cd.s/m2 light stimulus) are shown.
Blood collection and serum lactate analysis
Blood was collected from the retro-orbital plexus and serum was isolated by centrifugation. Lactate concentration was determined using a colorimetric assay following the manufacturer’s instruction (Abcam, ab65331). Measures of absorbance were done on a PHERAstar FS plate reader (BMG Labtech) and collected with the MARS data analysis software version 3.20 R2 (BMG Labtech). Hematocrit was determined by measuring the space occupied by red blood cells after centrifugation of blood in heparinized capillaries (Brand Micro-hematocrit capillaries 749311) at 10,000 rpm for 5 minutes.
BN-PAGE, total OxPhos immunostaining and complex I in-gel activity assay
The BN-PAGE and in gel activity assay protocols were described in detail previously(Jha et al., 2016). Briefly, ∼30 mg of frozen brain tissue were homogenized in ice-cold isolation buffer (0.2M sucrose, 10mM Tris, 1mM EGTA/Tris pH 7.4, adjust pH to 7.4 with 1M HEPES buffer, protease inhibitors). Following centrifugation, isolated mitochondria protein content was quantified using DC protein assay (Bio-rad). For BN-PAGE immunoblotting and in-gel activity, 50μg of mitochondria extract were solubilized using 5% digitonin. Electrophoresis of solubilized mitochondrial proteins was performed using the NativePAGE system (Novex) using 3-12% gradient gels. For immunoblotting samples were run at 150V for 30 min and at 250V for additional 90min. Proteins were transferred on a PVDF membrane using an iBlot Gel Transfer device (Invitrogen) and incubated with primary antibodies (Anti-OxPhos Complex Kit (Life technologies, cat. no. 457999) and Anti-MTCO1 antibody (Abcam, ab14705) to detect total OxPhos proteins. Immunostaining of the membrane was performed using Western Breeze Chromogenic Immunodetection System (Invitrogen). For complex I in-gel activity assay samples were run at 150V for 30 min and at 250V for additional 150min to obtain maximal separation of supercomplexes bands. Gels were run at 4°C to preserve enzymatic activity. After electrophoresis gels were incubated with Complex I substrate solution (2mM Tris-HCl, 0.1 mg/ml NADH, 2.5 mg/ml Nitrotetrazolium Blue chloride) at RT until the appearance of purple bands indicative of complex I activity (10-20 min).
Tissue staining and microscopy
Mice were deeply anesthetized by intra-peritoneal injection of ketamine and were then perfused through the ascending aorta with PBS. Tissues were dissected and post-fixed during 24 hours at 4°C in 4% paraformaldehyde. Tissues were cryoprotected overnight in 25% sucrose before being embedded in OCT and cut on a cryostat (Leica CM3050S or Thermo Fisher Scientific CryoStar NX50).
Immunohistological staining was performed as previously described(Decressac et al., 2012). Sections were washed with PBS, incubated in 0.1 M PBS containing 10% normal goat serum, 0.25% Triton X-100 and the primary antibodies for 16–18 hours at room temperature, shaking gently. The tissue sections were then washed three times in PBS and incubated in secondary antibodies secondary antibodies coupled to Alexa 488 or Alexa 568 (1:300, Molecular Probes, Invitrogen) diluted in the blocking solution for 2 hours at room temperature. Sections were rinsed again and coverslipped using the Vectashield Hard Set mounting medium (Vector Laboratories, Burlingame, CA) or the Prolong Gold Antifade medium (Molecular Probes).
Primary antibodies were used to detect GFAP (rabbit, 1:2000, Abcam, #ab7260), Iba1 (rabbit, 1:2000, Wako, #019-19751), GFP (chicken, 1:2000, Abcam, #ab13970), Calbindin (mouse, 1:1000, Millipore, #ab1778), NeuN (rabbit, 1:1000, Cell Signaling, #12943), TH (mouse, 1:3000, Immunostar, #22941).
Staining of F-actin on liver samples was performed by incubating sections with Alexa Fluor 594 Phalloidin (1:600; Molecular Probes, A12381) together with the secondary antibodies.
For the staining of lipid droplets, brain sections were rinsed in PBS and then incubated for 10 min in Nile red (2.5μg/ml dissolved in 75% aqueous glycerol). Subsequently, sections were washed twice with PBS, mounted and coverslipped with Prolong Gold Antifade medium (Molecular Probes). Images were captured using the LSM 710 confocal microscope (Zeiss).
Quantification of retinal ganglion cells and Purkinje cells
The number of retinal ganglion cells was determined by counting NeuN-positive cells in the retinal ganglion cell layer. The number of Purkinje cells was determined by counting calbindin-positive cells in the cerebellum. Images were analyzed using ImageJ v.2.0.0 software (National Institute of Health, Bethesda, MD, USA). Quantification was performed in 10 random fields of view per mouse and the average number of NeuN-positive or calbindin-positive cells per mm was calculated.
Analysis of cardiac pathology
Analysis of cardiac pathology was first determined by calculating the heart weight (g) / body weight (g) ratio. In addition, we performed a morphometric analysis of the cardiac tissue. 35μm-thick heart sections were cut on a cryostat (Leica CM3050S or Cryostar NX50) followed by staining with wheat germ agglutinin (WGA) and DAPI. Free floating sections were washed in HBSS and stained with WGA-CF™ 640 (5μg/ml, HBSS) (Biotium) for 4 hours at 37°C to visualize cell membranes. Nuclei were stained with DAPI (5μg/ml, PBS) (Invitrogen) for 5 min before being rinsed 3 times in PBS. Sections were mounted on glass slides and coverslipped with the ProLong Gold Antifade mounting medium (Molecular Probes). Images of entire sections were acquired using the Axio Scan.Z1 slide scanner (Zeiss) and images of the WGA staining were captured with a LSM 710 confocal microscope (Zeiss). For each animal, cardiomyocytes cross-sectional area was measured from 10 different pictures of WGA-stained sections. A minimum of 250 cells were analyzed using ImageJ v.2.0.0 software (National Institute of Health, Bethesda, MD, USA).
Western blotting
Tissues were harvested and homogenized in RIPA buffer (Sigma). 20 to 40μg of protein were boiled at 95°C for 5 minutes in Laemmli buffer (Biorad), separated on a SDS-PAGE gel and then electrotransfered (100V, 1 hour) on a PVDF membrane (Millipore). After blocking for 1 hour in Tris-buffered saline with 0.1% Tween-20 (TBS-T) and 3% non-fat dry milk, membranes were incubated overnight at 4°C with one of the following primary antibodies: GFP (chicken, 1:2000, Abcam, #ab290), GFAP (rabbit, 1:2000, Santa-Cruz, #sc-6170), Ndufs4 (mouse, 1:2000, Abcam, #ab87399), Calnexin (Rabbit, 1:5000, Enzo Life Science, ADI-SPA-860). After washing for 30 minutes in TBST, membranes were incubated for 1 hour at room temperature with an HRP-conjugated secondary antibody (1:1000, Promega). Protein expression was revealed using the Clarity kit (Biorad). Luminescence signal was detected using the GE Amersham AI600 (GE Healthcare) or the ChemiDoc MP (Biorad) and band intensities were quantified by densitometry using ImageJ v.2.0.0 software (National Institute of Health, Bethesda, MD, USA).
Statistical analysis
Statistical analysis was conducted using the GraphPad Prism software (version 7.0a). One-way ANOVA test with Tukey’s multiple comparison test were performed to analyze the difference between experimental groups. Student’s t-test was used to analyze the difference between two groups. The data were collected and processed in a randomized and blinded manner. No statistical methods were used to predetermine sample size, but our sample sizes are similar to those generally employed in the field. All values are presented as mean ± standard error of the mean (SEM). Statistical significance was set at P<0.05.
Data availability
All the data and reagents (if not commercially available) that support the findings of this study are available from the corresponding author upon reasonable request.
Results
Ndufs4 KO mice present with severe pathology in several organs including the brain, the retina, the liver and the heart (Kruse et al., 2008; Karamanlidis et al., 2013; Chouchani et al., 2014; Jin et al., 2014; Yu et al., 2015). We first verified that the AAV.PHP.B vector was efficient in transducing these tissues following intravenous injection in 1-month-old mice. Two weeks after AAV vector injection, transgene expression was detectable throughout the central nervous system (Supplementary Fig. 1a). GFP-positive cells were observed in the olfactory bulb, striatum, hippocampus, midbrain, cerebellum, spinal cord and vestibular nucleus and co-localized with ubiquitous or subtypes-specific neuronal markers (Supplementary Fig. 1b-c). The AAV-PHP.B vector was also efficient in inducing transgene expression in other tissues including the liver, heart, muscle and retina (Supplementary Fig. 1c-d).
These results prompted us to examine the effect of systemic delivery of an AAV-PHP.B-Ndufs4 vector (thereafter referred as AAV-Ndufs4 vector) in the Ndufs4 KO mouse model of LS. One-month-old KO and control mice were randomly assigned to receive an intravenous injection of an AAV-Ndufs4 or a GFP-expressing control vector (1 × 1012 genome copies/mouse) and they were regularly monitored throughout their survival period. In line with previous reports, KO mice treated with the control GFP vector died at a median age of 54 days. In contrast, gene replacement therapy substantially increased the lifespan of LS mice as 50% of the KO treated with the AAV-Ndufs4 vector survived 250 days (log-rank P<0.0001; χ2=38.36) (Fig. 1a). Bi-weekly monitoring of the body weight of the animals showed that delivery of the Ndufs4 gene improved this parameter and promoted a continuous, yet slightly delayed, growth of KO mice (Fig. 1b-c). At 5-6 months of age, AAV-Ndufs4-treated KO mice were barely distinguishable from control animals (P=0.23 at 170 days) (Fig. 1b). Before dying, the body temperature of Ndufs4 KO mice falls progressively below 34°C (Fig. 1d) (Jain et al., 2016). Gene replacement therapy rescued hypothermia as KO animals treated with the AAV-Ndufs4 vector maintained a core body temperature similar to those of control mice (Fig. 1d) (P>0.05). We confirmed that these physiological improvements were associated with a restoration of Ndufs4 expression in the brain, liver, heart and muscle (Fig. 1e). Complex I is essential for the assembly of mitochondrial super-complexes (Moreno-Lastres et al., 2012). Consistent with the recovery of Ndufs4 expression, blue native in-gel activity assay and total OxPhos immunostaining demonstrated the physical restoration of complex I and super-complexes as well as an improvement of mitochondrial complex I activity (Fig. 1e; Supplementary Fig. 2).
As LS primarily affects the brain and causes severe neurological symptoms in patients, we first assessed the paw clasping phenotype in mice (Johnson et al., 2013). While 75% of AAV-GFP treated KO mice displayed clasping behavior at 45 days of age, only 13% showed this phenotype in the AAV-Ndufs4 treated group (Fig. 2a). Notably, none of the KO mice treated with the AAV-Ndufs4 vector showed this phenotypic sign at 250 days of age (Fig. 2a). In addition, the cylinder test revealed a long-lasting improvement of locomotor activity in the AAV-Ndufs4 treated group as seen by a reduction in the time spent inactive and a higher number of rears compared to KO mice treated with the control vector (Fig. 2b-d and Supplementary video 1-3). Likewise, long-term improvement in strength and resistance were observed in the hanging test (Fig. 2e). Brain pathology was also illustrated by the occurrence of epileptic seizures in about 23% of KO mice treated with the GFP-expressing viral vector (Fig. 2f and Supplementary Video 4) (Quintana et al., 2010). In contrast, only 5% of KO mice injected with the AAV-Ndufs4 vector elicited epileptic episodes (Fig. 2f).
As seen in LS patients, Ndufs4 KO mice show necrotic plaques and severe neuroinflammation in various brain regions including the cerebellum, the vestibular nucleus and the olfactory bulb (Kruse et al., 2008; Johnson et al., 2013). In line with our observation that the AAV-PHP.B vector is efficient in infecting these brain areas and provides improvement in behavioral performances, western blot and histological analysis of the microglia (Iba1) and astrocyte (GFAP) markers confirmed the presence of severe gliosis in the brain of 45 day-old KO mice and revealed that AAV-mediated gene therapy prevented the development of this inflammatory reaction (Fig. 3a-d). In parallel, lipid accumulation in glial cells resulting from the neuronal metabolic distress was markedly reduced at 45 and 250 days of age in the brain of mice treated with the AAV-Ndufs4 vector (Fig. 3e) (Liu et al., 2015; Liu et al., 2017). To assess the effect of gene replacement on neurodegeneration, we quantified the number of calbindin-positive Purkinje cells in the cerebellum and we found that AAV-Ndufs4 vector delivery rescued a large number of these neurons in KO mice at both 45 and 250 days (Fig. 3f-g) (Quintana et al., 2010).
In addition to the brain, the retina is severely affected in the LS mouse model (Yu et al., 2015; Song et al., 2017). To test whether restoration of Ndufs4 expression in the retina by gene delivery translated into improved retinal function, we performed electroretinogram recordings and demonstrated that restoration of Ndufs4 expression was associated with the maintenance of the B-wave response that was absent in KO mice (Fig. 4a-b). In line with these results and with our observation that peripheral delivery of the PHP-B vector efficiently transduced the inner retina including the ganglion cell layer and the inner nuclear layer (Supplementary Fig. 1d), we showed that the injection of the AAV-Ndufs4 vector in Ndufs4 KO mice protected NeuN-positive retinal ganglion cells as compared to the KO+GFP group (P<0.001) (Fig. 4c-d).
Besides the brain pathology, Ndufs4 KO exhibit peripheral symptoms including metabolic perturbations (Johnson et al., 2013; Jin et al., 2014; Jain et al., 2016). We observed that gene replacement was effective in normalizing the polycythemia (Supplementary Fig. 3a) and in lowering the level of blood lactate in Ndufs4 KO mice (Supplementary Fig. 3b). In addition, it has been demonstrated that deletion of the Ndufs4 gene drives hypertrophic cardiomyopathy (Karamanlidis et al., 2013; Chouchani et al., 2014). Consistent with our observation that systemic delivery of the AAV-PHP.B vector efficiently targeted the cardiac tissue (Supplementary Fig. 1c), we showed that treatment with the Ndufs4-expressing vector corrected this cardiac abnormality both at the anatomical and cellular level (Supplementary Fig. 3c-f).
Discussion
Our study provides a proof-of-principle that gene replacement can exert a robust therapeutic effect in an adult mouse model of LS. These results have two important implications. We report the first-ever successful gene therapy with long-term benefits in this pre-clinical model of severe mitochondrial disease. This is an important step considering that previous attempts resulted in marginal extension of lifespan. Indeed, targeted delivery of a Ndufs4-expressing AAV vector to the vestibular nuclei or combined systemic and intracerebroventricular injections of a similar viral vector in neonates slightly increased the median survival to 65 and 82 days, respectively (Quintana et al., 2012; Di Meo et al., 2017). In contrast, Ndufs4 KO mice treated with rapamycin or exposed to hypoxic conditions survived more than 100 days (Johnson et al., 2013; Johnson et al., 2015; Jain et al., 2016; Ferrari et al., 2017). Our results indicate that widespread gene replacement improves survival to a degree at least as good as the ones reported for these non-genetic strategies. Although we performed a single-dose study, one can anticipate that lower doses of AAV.PHP.B vector would afford milder therapeutic effects. This assumption is based on the results from both the AAV-based strategy and the non-genetic approaches which showed that reduced dosage or intermittent treatment have reduced efficacy (Johnson et al., 2013; Johnson et al., 2015; Jain et al., 2016; Di Meo et al., 2017; Ferrari et al., 2017). It is essential to emphasize that gene therapy should remain the ultimate goal for such genetic condition as it is the only strategy that can permanently and efficiently fix the primary defect. Pharmacological or other non-genetic approaches could nevertheless serve as adjunctive treatments to alleviates residual defects.
Regardless of the clinical relevance of the strategy used, it is fundamental to first report that a fatal disease such as LS can be rescued. This goes along with the identification of the organs driving the main pathological manifestation. Despite the conditional ablation of Ndufs4 in the brain clearly established that the CNS plays a critical role in LS pathology (Kruse et al., 2008; Quintana et al., 2010), the demonstration that reinstating its expression is feasible and provides a therapeutic benefit had to be made. Similarly reports have been important steps for other rare diseases with multi-organ deficiency such as lysosomal diseases or multiple system atrophy and have paved the road for future clinical development (Hua et al., 2011; Spampanato et al., 2011). The clinical transferability of the tools, methods and doses used to make this first demonstration are questions that need to be addressed in future experiments. For our study, the dose (1 × 1012) was selected based on previous reports describing widespread and strong transgene expression in mice (Deverman et al., 2016; Challis et al., 2019). Although recent studies have shown that the AAV.PHP.B vector produced a different and more restricted pattern of expression in non-human primates as compared to mice, this new AAV variants remains a powerful tools for initial demonstration of therapeutic rescue as reported here (Hordeaux et al., 2018; Liguore et al., 2019).
The second important information provided by our work is that delayed intervention can still rescued the disease phenotype even if the therapeutic window to intervene is narrow as it is the case for pediatric conditions like LS. Despite the pathology is already prominent in one-month old Ndufs4 KO mice, we decided to administer the AAV vector at a stage where most organs are almost fully mature hence reducing the risk of a time-dependent decrease in transgene expression (Di Meo et al., 2017). This timeline is also clinically relevant since children are diagnosed after the presentation of symptoms.
Beyond the findings reported here in a mouse model of LS, our study brings promising perspectives for other fatal conditions with CNS involvement as well as for less severe pathologies with later onset.
Authors contribution
R.R.D, G.B, E.M, R.A-T, and M.D performed the experiments, R.R.D, G.B, E.M, E.M.S and M.D prepared figures, R.R.D, G.B, E.M.S, J.A and M.D designed experiments, R.R.D, G.B, E.M, E.M.S, J.A and M.D wrote the manuscript, R.R.D, G.B, E.M, E.M.S, J.A and M.D analyzed the results.
All authors discussed the results and commented on the manuscript at all stages.
Supplemental information
Supplementary Video 1: Cylinder test performed at 45, 100 and 250 days by the same mouse from the Ctl+GFP group.
Supplementary Video 2: Cylinder test performed at 45, 100 and 250 days by the same mouse from the KO+GFP group.
Supplementary Video 3: Cylinder test performed at 45, 100 and 250 days by the same mouse from the KO+Ndufs4 group.
Supplementary Video 4: Epilepsy seizure observed in a 45 days-old KO mouse treated with the AAV-PHP.B-GFP vector.
Acknowledgments
We thank Edoardo Nusco, Mathilde Faideau, the TIGEM AAV Vector Core, and the TIGEM animal facility staff for excellent technical assistance. We thank the Photonic Imaging Center of Grenoble Institute Neurosciences (Univ. Grenoble Alpes – Inserm U1216) which is part of the ISdV core facility and certified by the IBiSA label. We are grateful to Pr. Richard D. Palmiter (University of Washington, Seattle) for providing the pAAV-CBA-Ndufs4-IRES-GFP plasmid and to Dr. Viviana Gradinaru for providing the PHP.B packaging plasmid.
This work was supported by the Fondazione Telethon (grant #TMMDMTX16TT to M.D), by the Agence Nationale de la Recherche (ANR-JCJC program, grant #ANR-17-CE37-0008-01 to M.D), by the grants from the Agence Nationale de la Recherche ANR-15-IDEX-02 NeuroCoG (to M.D) in the framework of the “Investissement d’Avenir” program, by the European Research Council (ERC-AdG-787702 “UPRmt” to J.A, and ERC.311682 “Allelechoker” to E.M.S), by the Swiss National Science Foundation (SNSF 310030B-160318 to J.A), by the Strategic Focal Area “Personalized Health and Related Technologies (PHRT)” of the ETH Domain (grant #2018-422 to G.B), by a GRL grant of the National Research Foundation of Korea (NRF GRL 2017K1A1A2013124 to J.A) and by the EPFL. M.D is supported by an IDEX Chair of Excellence from the University of Grenoble-Alpes and by the Edmond J. Safra Foundation. M.D laboratory is member of the Grenoble center of Excellence in Neurodegeneration. The authors also want to thank the Fondation Bettencourt Schueller and the Plan Cancer.
Authors declare that no competing financial and/or non-financial interests exist.