RNA-seq, proteomics, flux proteomics, unlabeled metabolomics and C13-glucose metabolomics of Lamp2aWT and Lamp2aPax7CreER Muscle Stem Cells

Description

Series information

RNA-seq experiment (CMA90) is encoded in Supplementary Files 6

Samples: MuSCs from Lamp2aWT and Lamp2aPax7CreER mice were isolated from uninjured (0), and injured with cardiotoxin (CTX) at 1 or 4 days post-injury (dpi). MuSCs were isolated by FACS-sorting (CD31-, CD45-, Sca1-, a7-integrin+ and CD34+). MuSCs, from each condition, were divided into quadruplicates and subjected to bulk RNA-seq --> 24 samples in total (4 * 2 genotypes * 3 timepoints)

Methods: 

Sample and library preparation for RNA-seq. Sequencing libraries were prepared directly from the lysed cells, without a previous RNA-extraction step. RNA reverse transcription and cDNA amplification were performed using the SMART-Seq v4 UltraLow Input RNA Kit for Sequencing (Clontech Takara, # 634894). The IlluminaNextera XT kit was used to prepare the libraries from the amplified cDNA, and libraries were sequenced using the HiSeq 2500 (Illumina) (51 bp read length, single-end, around 20 million reads).

Bulk RNA-seq data processing. Obtained reads were pre-processed using a custom pipeline that assesses read quality with FastQC (Babraham Institute, https://www.bioinformatics.babraham.ac.uk/projects/fastqc/), and uses Cutadapt⁴¹ to trim Illumina adaptors and reads shorter than 30 bp. Filtered reads were mapped against the mouse reference transcriptome (GRCm38) and RSEM⁴² tool was used to calculate gene expression levels as raw counts and normalized counts (norm. counts).

Differential gene expression and pathway enrichment analysis. Only genes with more than 4 counts in all but one sample in at least one experimental group were considered for the analysis. We considered only the genes with more than 4 CPM in all but one sample in at least one experimental group. Differentially expressed genes (DEGs) were obtained with Bioconductor package limma⁴³ applying voom transformation, robust regression method for linear modeling, and empirical Bayes statistics. Only genes with adjusted p-value < 0.05 were considered for further analysis. For pathway enrichment analysis⁴⁴, DEGs were analyzed on g:Profiler platform (https://biit.cs.ut.ee/gprofiler/gost). Total annotated genes from the full reference genome were used as the reference set and pathway significance was set by applying a Benjamini-Hochberg FDR method.

Methods

Proteomic flux experiment (CMA135) is encoded in Supplementary File 7

Samples: Cultured MuSCs from Lamp2aWT and Lamp2aPax7CreER mice were treated with lysosomal inhibitors (20 mM NH₄Cl and 100 μM leupeptin (NL)) for 18h. 1 million MuSCs were isolated and whole-cell protein extraction was performed for subsequent flux proteomics analysis. --> n = 3 Lamp2a^WT Ctrl, n = 3 Lamp2a^WT Leu, n = 3 and Lamp2a^ΔPax7ER Ctrl, n = 2 and Lamp2a^ΔPax7ER Leu, total 11 TMT channels.

Methods: Proteins were obtained from cultured Lamp2a^WT and Lamp2a^ΔPax7ER MuSCs treated with and without Leupeptin 100 μM (Sigma, # L2884-10MG) for 18h. Protein samples were extracted by boiling in the presence of an SDS-containing extraction buffer, and protein concentration in the resulting preparations was determined using the RCDC Protein Assay Kit (Bio-Rad, Hercules, CA, USA). Protein tryptic digestion was carried out using 30 kDa FASP filters (Abcam, Cambridge, UK) following the manufacturer’s instruction⁴⁹, after which the resulting peptides were isobarically labeled with 11-plex Tandem Mass Tags (TMT, Thermo Scientific, Waltham, MA USA) reagents following the manufacturer’s instructions⁵⁰ into two TMT replicates containing samples from all conditions (3 Lamp2a^WT Control, 3 Lamp2a^WT Leu, 3 Lamp2a^ΔPax7ER Control and 2 Lamp2a^ΔPax7ER Leu). Liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis was performed on an Easy-nLC 1000 HPLC system (Thermo Fisher Scientific) coupled via a nanoelectrospray ion source (Thermo Fisher Scientific) to a Fusion tribrid mass spectrometer (Thermo Fisher Scientific) as described elsewhere, and the raw LC-MS/MS data were searched against a Uniprot Mus musculus database (March 2022; 65,300 entries) with Proteome Discoverer (version 2.5, Thermo Fisher Scientific) for peptide identification^52,53. Statistical analysis of protein and peptide abundance changes was carried out using the iSanXoT software package^54–56. Relative protein abundances are expressed as standardized log2 (Fold Changes) (Zq). Finally, the limma package⁴³ was used to ascertain statistical significance employing p-values.   

Methods

Proteomic experiment (CMA213) is encoded in Supplementary File 8

Samples: Cultured MuSCs (300.000 cells/condition) from Lamp2aWT and Lamp2aPax7CreER mice were isolated and whole-cell protein extraction was performed for subsequent proteomics analysis --> 12 samples (2 genotypes * 6 replicates)

Methods: Cultured Lamp2a^WT and Lamp2a^ΔPax7ER MuSCs were collected, lysed and homogenized in 2.5% SDS in 100 mM Tris pH 8.1 at 95°C for 5 min. Lysates were clarified by centrifugation for 5 min at 30,000x g. Protein concentrations were determined using the BCA assay and 100 µg of protein was precipitated with four volumes of ice-cold acetone and resuspended at 5 g/l in 100 mM Tris pH 8.1 containing 8 M Urea. Proteins were digested with LysC at 1:50 enzyme:protein ratio for 6 h at 25 °C, 1000 RPM. Digests were diluted with four volumes of 100 mM Tris pH 8.1, and trypsin was added to the samples at 1:50 enzyme:protein ratio, digestion continued for a further 16 h. Following digestion, samples were reduced by addition of DTT to 1 mM for 15 min at 25 °C, and alkylated by addition of iodoacetamide to 5 mM for 15 min at 25°C in the dark. Samples were acidified to 1% trifluoroacetic acid (TFA). Prior to peptide desalting, digests were filtered through a 0.2 µm filter (Part No.) by centrifugation at 1000 g for 1 min, according to the manufacturer´s instructions. Briefly, the filter plate was activated with 400 µl of methanol and equilibrated with 400 µl of 1% TFA. 250 µl of sample was loaded and combined with an additional 250 µl of 1% TFA wash. Combined flowthrough was then subjected to solid-phase extraction on an Agilent AssayMap Bravo using 5 µl C18 cartridges. Cartridges were activated with 50% ACN, 0.1% TFA, equilibrated with 0.1% TFA, samples loaded, washed with 0.1% TFA then eluted with 70% ACN, 0.1% TFA in water. Desalted peptide mixtures were centrifuged under vacuum, and subsequently resuspended with 3% ACN, 0.1% FA in water, at a concentration of 2.5 g/l.

For LC-MS/MS analysis, a ThermoFisher Vanquish Neo UHPLC was coupled to a ThermoFisher Orbitrap Ascend Tribrid Mass Spectrometer. All results were obtained on a Waters Acquity CSH C18 column (1.7 µm, 1 mm X 150 mm, Part No). Column oven temperature was maintained at 55°C. Peptides were separated using solvent A: 0.1% FA in LCMS water, solvent B: 80% ACN, 0.1% FA in LCMS water, on a two-step linear gradient of 3 – 35% B in 56 min, followed 35 – 45% B in 4 min, at a constant flow rate of 50 μL/min. Ions were generated using the heated electrospray ionisation source, with positive spray voltage of 3.5 kV, capillary temperature of 325°C, vaporizer temperature of 125°C. The flow rates for sheath gas, aux gas, and sweep gas were 32, 5, and 0 respectively. Full MS scans, from 350-1050 m/z were acquired in the Orbitrap analyzer at resolution of 120,000, normalized AGC Target was 100%, with maximum injection time set to 251 ms. RF lens % was 60. Seventy partially overlapping 9.5 m/z quadrupole windows covering the mass range 384.5-1015 were used to isolate an AGC target of up to 1000% in a maximum of 27 ms, and subjected to HCD, with normalized collision energy at 31%, with measurement in the orbitrap at a resolution of 15,000, using the scan range 200 – 1800 m/z. The resulting spectra were searched using DIA-NN Software v.1.8 with an in silico predicted library of the Mouse proteome from SwissProt.

Methods

Unlabeled and C-13 metabolomic experiment (CMA213 and CMA214) are encoded in Supplementary Files 9 and 10

Samples: Cultured MuSCs (300.000 cells/condition) from Lamp2aWT and Lamp2aPax7CreER mice were isolated and whole-cell metabolite extraction was performed for subsequent metabolomic analysis --> 12 samples (2 genotypes * 6 replicates)

Methods:

Sample preparation and analysis for unlabeled or C¹³-glucose-labeled metabolomics. For unlabeled metabolomics, 300,000 cells per well were grown in 6-well plates for 18 h prior to collection. For C¹³-glucose-labeled metabolomics, 200,000 cells per well were grown in 6-well plates for 24 h and then incubated for 1 h in non-glucose GM:DMEM without glucose (Gibco, 11966025), 20% dialyzed FBS (Cytiva, SH30079.03IR25-40), 1% sodium pyruvate (ThermoFisher, 11360039) and bFGF (ThermoFisher, 100-18B-50UG). Next, GM was replaced with GM with unlabeled glucose (Agilent, 103577-100) or C¹³-labeled glucose (Cambridge Isotope Laboratories, CLM-1396-1) at 100 mM, and cells were collected at 0, 5, 15, 30, 60 and 240 min after glucose addition. To collect samples, plates with seeded cells were washed twice with ice-cold saline, flash frozen in liquid nitrogen, and quenched on ice in 700 μL of extraction buffer 1 (80:20 MeOH:H₂O) containing internal standards. The collected extract was mixed with 800 μl of Extraction Buffer 2 (53 MeOH, 34 CHCl₃, 12 H₂O, 1 CH₃COOH) and transferred into a 2.0 ml Eppendorf tube. Samples were shaken on a ThermoMixer (Eppendorf) at 1200 RPM for 30 min at 4°C and centrifuged at 30,000 g for 10 min at 4°C. The aqueous layer was transferred to new Eppendorf tubes and dried overnight on a Labconco speedvac at 4°C. Once dried, samples were resuspended in 100 μl of Resuspension Buffer (40 ACN : 40 MeOH : 20 H2O). Metabolites were quantified using an Exploris 240 mass spectrometer (Thermo Fisher Scientific) connected to a Vanquish Duo UHPLC chromatography system (Thermo Fisher Scientific) in both positive and negative mode. Anionic metabolites (Negative Mode) were separated on an Acquity PRM GLY BEH 130A 1.7 μm 2.1 x 150   (Waters #186009524) column at a flow rate of 0.4 ml min^–1 using a gradient consisting of Mobile Phase A (10 mM ammonium acetate in a solution of 90 H2O : 10 ACN and 5 μM medronic acid) and Mobile Phase B (10 mM ammonium acetate in 80 ACN : 10 H2O and 5 μM medronic acid). In brief, the gradient started at 0% Mobile Phase A, before ramping to 30% Mobile Phase A over 8 min and continuing at 50% A for 1 min. At 9:50 (min:seconds), the gradient stepped down to 0% A for 50 seconds, before ramping and holding at 0% A for 3 min. The column temperature was set at 40°C. Cationic metabolites (Positive Mode) were separated on a HILIC Z: InfinityLab Poroshell 120 HILIC-Z, 2.1 x 150 mm, 2.7 µm (Agilent technologies #683775-924) column at a flow rate of 0.40 ml/min with a gradient composed of Mobile phase A (10 mM ammonium formate in 0.1% formic acid and water) and Mobile phase B (0.1% acetonitrile). In brief, the gradient started at 5% A for 1 min, before ramping to 50% A over 6 min. For 3 min, the gradient was maintained at 50% A, before stepping down to 5% A for 2 min. The column temperature was set at 40°C. A full mass scan was performed with the following settings: range, 70 to 900 m/z; resolution, 120,000; automatic gain control target, Standard; and maximum ion injection time, Auto. The ion source parameters were as follows: spray voltage, Static; transfer temperature, 275°C; RF-lens level, 75; heater temperature, 320°C; sheath gas, 50; and aux gas, 10. Raw data was processed using SciexOS (AB Sciex) and Skyline⁵⁷.

Metabolomic data analysis. Chromatographic peak integration and extraction from raw data was performed in a targeted manner using Skyline⁵⁷. Metabolites with poor peak shapes were excluded. Analytes (excluding spiked internal standards) were median-scaled across samples. The median of these median-scaled analyte values across all analytes was used for each sample as the normalization factor. Each analyte measurement value was divided by this normalization factor to compute the final normalized value. QC replicates from pooled samples were interspersed with study samples. The one with the lowest QC variability (CV < 25%) was retained for metabolites measured by multiple methods. For the isotope tracing experiment, data were collected on two sets of samples for each experimental group. Unlabeled samples were corrected for natural isotopes using IsoCorrectoR⁵⁸. Isotopologues with adjusted signals > 0.5 % of M0 were removed. Among the LC-MS methods, the one with the fewest non-target signals (or the highest signal if tied) was chosen. For PCA analysis, isotopologue values were normalized to proportions, center log-ratio transformed (after imputing missing values via cmultRepl function from zCompositions package⁵⁹), Z-score transformed and analysed using PCA (prcomp). For downstream analysis, the percentage of labeled isotopologues was considered as a proxy of the amount of metabolic labeling of a given metabolite.