Impairment of the mitochondrial one-carbon metabolism enzyme SHMT2 causes a novel brain and heart developmental syndrome

Àngels García‐Cazorla1,2 · Edgard Verdura2,3 · Natalia Juliá‐Palacios1,2 · Eric N. Anderson4 · Leire Goicoechea2,3 · Laura Planas‐Serra2,3 · Enkhtuul Tsogtbaatar5,6,7 · Nikita R. Dsouza8 · Agatha Schlüter2,3 · Roser Urreizti2,9 · Jessica M. Tarnowski10,11 · Ralitza H. Gavrilova10,11 · SHMT2 Working Group · Montserrat Ruiz2,3 · Agustí Rodríguez‐Palmero2,3,12 · Stéphane Fourcade2,3 · Benjamin Cogné13 · Thomas Besnard13 · Marie Vincent13 · Stéphane Bézieau13 · Clifford D. Folmes5,6,7 · Michael T. Zimmermann8,14 · Eric W. Klee10,15,16 · Udai Bhan Pandey4 · Rafael Artuch2,9 · Margot A. Cousin15,16 · Aurora Pujol2,3,17


Molecular Dynamics Simulations
Similar to previous works [13,30,31], we carried out Molecular Dynamics (MD) using the Generalized Born implicit solvent molecular dynamics (isMD) simulations, which were in NAMD and the CHARMM36 [14] force field. The homology model was utilized for our initial conformation for the WT. In silico mutagenesis was performed using FoldX for the mutations and gnomAD variants. We utilized an interaction cutoff of 12Å with strength tapering (switching) beginning at 10Å, a simulation time step of 1fs, conformations recorded every 2ps. Each initial conformation was used to generate 3 replicates, and each replica was energyminimized for 10000 steps, followed by heating to 300K over 300ps via a Langevin thermostat. A further 12ns of simulation trajectory was generated and the final 10ns were analyzed. All trajectories were first aligned to the initial wild type conformation using Cα atoms. Root meansquared deviations (RMSD) and root mean-squared fluctuations (RMSF) were calculated using Cα atoms. Principal Component (PC) analysis was performed in Cartesian space. Analysis was carried out using custom scripts, leveraging VMD and the Bio3D R package. We used the approach of Karamzadeh et al [16] to calculate Free Energy Landscapes (FELs) from PC sampling. Protein structure visualization was performed with PyMol and VMD.

Fibroblast cultures
Fibroblast cell lines from the skin biopsy samples of Patients 1-5 and controls were cultured at 37ºC and 5% CO 2 in Dulbecco's modified Eagle's Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 100 U/ml penicillin and 100 μl/ml streptomycin. Fibroblast passages ranged from to 10 to 15 in all experiments, and controls were age-matched individuals.

Mitochondrial enrichment
Fibroblast pellets were homogenized in 4 volumes of ice-cold homogenization buffer (HB) with a 1 mL needle at 4ºC. The HB was prepared immediately before use, and it was composed of 225 mM D-mannitol, 25 mM HEPES-KOH, 1 mM EGTA, and protease inhibitors, pH 7.4. The remaining homogenate was centrifuged at 600 x g for 10 min at 4 ºC to precipitate the nuclei and unbroken cells. The supernatant was centrifuged again under the same conditions to eliminate all debris, and was then centrifuged at 10000 x g for 20 min at 4ºC. The supernatants were enriched in ER, cytosol, lysosomes, and depleted of mitochondria, and were kept separated. The pellets comprised the crude mitochondria or enriched mitochondria fraction.

RNA extraction and quantitative real-time PCR
RNA was extracted from fibroblasts using an RNeasy Mini kit (QIAGEN) and cDNA was synthesized using a Superscript IV kit (Life Technologies) following the manufacturers' instructions. RT-PCR primers to sequence cDNA in Patient 3 were: SHMT2_F: GAGGACCGAATCAACTTTGC and SHMT2_R: CTCACGGAACTGTCGAGAAG

ATP measurement
Fibroblasts were plated in 96-well transparent and white plates in parallel. The transparent plates were used to monitor cell growth and for cell nuclei quantification. The white plates were used to measure luminescence. Cells were deprived of glucose for 24 h using DMEM medium without glucose (Biological Industries, ref: 01-057-1A). Then, ATP concentrations were quantified using the ATPlite 1 step assay (PerkinElmer, 6016736) according to the manufacturer's protocol in a VICTOR X5 2030 Multilabel plate reader. Experiments were performed twice, using quatriplicate wells. ATP measurements were normalized to number of cells per well, estimated by counting cell nuclei stained with DAPI.

Inner mitochondrial membrane potential quantification
The inner mitochondrial membrane potential was measured as previously described [21]. Briefly, flow cytometry was performed using the voltage-sensitive indicator, tetramethylrhodamine ethyl ester (TMRE) (Molecular Probes, T669). Fibroblasts were plated in 6-well tissue culture plates. When confluence was reached, cells were washed with PBS and incubated with 50 nM of TMRE in prewarmed PBS for 30 min at 37ºC. The cells were trypsinized, centrifuged at 1000 x g for 5 min and resuspended in prewarmed PBS. All samples were captured in a FACS CantoTM recording 20,000 cells for each condition and genotype tested. FCCP (Sigma, C2920) (200 µM for 10 min) was used as positive control. Histograms showing the percentage of depolarized cells were obtained after gating live cells. The data were analyzed with FlowJo Tree Star software. Experiments were performed twice.

Evaluation of intracellular ROS levels
Intracellular H 2 O 2 levels were estimated using the ROS-sensitive H 2 DCFDA (DCF) (Thermo Fisher, D399) probe. Intracellular and mitochondrial superoxide O − 2 anion levels were estimated using DHE and MitoSOX TM Red probes (Thermo Fisher, D11347 and M36008), respectively, as previously described [22]. The fluorescence of the DCF-, DHE-and MitoSOX-stained cells was measured with a spectrofluorometer (FLUOstar Omega microplate reader, BMG Labtech) (excitation wavelength 493 nm, emission wavelength 527 nm for DCF; excitation wavelength 530 nm, emission wavelength 590 nm for DHE and MitoSOX). For normalization, the protein levels were measured using a Pierce BCA Protein Assay protocol (Thermo Fisher Scientific). Antimycin A (Sigma, A8674) (Ant A at 200 µM for 1 h) was used as positive control. Experiments were performed twice.

Oxygen consumption measurements
The oxygen consumption rate (OCR) of the cells was measured using a Seahorse XFe96 Extracellular Flux Analyzer (Agilent Technologies) as previously described [23]. Briefly, fibroblasts were seeded into gelatin-coated cell culture microplates at a density of 70,000 per well and allowed to adhere for 24 h. Prior to the assay, the cell culture medium was replaced with XF DMEM medium (pH=7.4) containing 2 mM glutamine, 1 mM pyruvate, 5 mM HEPES and 1 mM glucose. The OCRs were measured after sequential injections of oligomycin (1µM), FCCP (2µM) and rotenone and antimycin A (1µM). For normalization, Hoechst 33342 stain was coinjected with rotenone and antimycin A at a concentration of 20 µM. The stained nuclei were counted using Cytation1 (BioTek) and used to normalize the OCRs. The average values (±SD) of the OCRs for basal respiration, maximal respiration, spare capacity and ATP coupling were calculated using the Mitochondrial Stress Test Export Function of the Wave software. The values are expressed as the mean±SD and statistical analysis was performed using Prism 8. Experiments were performed twice.

Drosophila lines and Motor function assay
Fly stocks and crosses were maintained on standard cornmeal agar mediate at 29° C. OK371-gal4 was acquired from Bloomington Stock Center and the Shmt2 (RNAi #1, ID: 19208; RNAi #2, ID: 19206) was acquired from Vienna Drosophila RNAi Center.OK371-gal4 (a motor neuron specific driver) was used to express Shmt2 RNAi or eGFP (control), and climbing assays were performed as previously described [1,7]. The climbing distance (cm) and velocity (cm/s) of each fly were quantified and analyzed using GraphPad Prism 6 software. Three experimental replicates were performed for each group.

Larval preparation, immunohistochemistry and quantification
For an analysis of neuromuscular junctions (NMJ), eGFP and Shmt2 (RNAi #1 and RNAi #2) wandering third instar larvae were dissected, fixed, and immunostained [7]. Briefly, the larvae were dissected in PBS and fixed in 4% paraformaldehyde in PBS for 20 min at room temperature, washed 3 times with 0.1% PBST (0.1% Triton X-100 in PBS), and then blocked with 5% normal goat serum (Abcam; AB7681) in 0.1% PBST. Larvae were then probed overnight with primary antibody mouse anti-DLG (DSHB, 4F3, 1:100) at 4°C. The larvae were then washed three times in 0.1 % PBST and incubated in secondary antibodies (anti-mouse Alexa Flour 647, Invitrogen, # 28181, 1:100) and Cy3-conjugated anti-HRP (Jackson ImmunoResearch, Cat#:123-165-021, 1:200) for 2 h, washed three times in 0.1% PBST, and mounted using DAPI Fluoroshield (Sigma, #F6182). Images were captured using Nikon A1 eclipse Ti confocal microscope. For the analysis, NMJs from muscle 4 on segment A2-A3 were imaged from 4-5 larvae, and then, the synaptic boutons were quantified using ImageJ software (NIH). Boutons included in a chain of two or more boutons were considered mature, while single boutons that were not included in a chain and were sprouted off of a mature bouton or branch were considered as satellite boutons. Statistical analysis was performed with Prism 6 (GraphPad Software).

19.
Lek M, Karczewski  Patient 4 (Family 3) is a female, first child of healthy Caucasian non-consanguineous parents, born at 40 weeks of gestation with microcephaly (31.5 cm; -2.3 SD), which had been detected during the third trimester of pregnancy. She said her first words and walked independently at 14 months and 24 months of age, respectively. At 2 years of age, she developed intention hand tremor, which was more marked at the left side, and pyramidal signs in the lower extremities (hypertonia, brisk deep tendon reflexes, and Babinski sign). Hypertrophic cardiomyopathy was detected at 9 years of age in the context of an infectious respiratory event. Severe dysfunction of the left ventricle (EF: 35%) led to heart transplantation some months later because of a rapid worsening (EF: 21%). The child is currently 12 years old, with microcephaly below the first percentile (-2.6 SD) and has moderate intellectual disability. She has a paraparetic genu-flexum gait but can walk long distances without help, mild intention tremor and dysmetria.
Patient 5 (Family 4) is a 2-year-old boy, the second child of healthy Caucasian nonconsanguineous parents. Microcephaly in the 3 rd centile was diagnosed prenatally, and a fetal cerebral MRI showed a two-week delay in maturation. He was born at 38 weeks of gestation with microcephaly (OFC <3 rd centile). He had early feeding difficulties, which was remediated by a nasogastric tube during the first weeks. Microretrognathism, bivid uvula, long palpebral fissures and overlapping sutures were present at birth. Cerebral computed tomography ruled out craniosynostosis. Echocardiography showed an atrial septal defect and ostium secundum. He has a severe global developmental delay, first holding his head up at 12 months, and unable to sit or speak at 2.5 years of age. He had no facial movement and showed hypersalivation, suggesting facial diplegia. He fails to thrive, with a weight at 2.5 SD, a height at -2 SD, and microcephaly with OFC at -3 SD.     In the case of the p.Gly375del model below (soft pink), the structure is superimposed on a cyan wild-type monomer. The deletion of Gly375 causes disruption of wild-type secondary structure, resulting into a different spatial positioning of Ser374 and Asp378, which likely interact with Arg230 in the wild-type situation. That would destabilize the loop that contains Arg230, which makes up part of the active site surface.