Study design
This study aimed to evaluate whether organelle inheritance controls CD8+ T cell differentiation. To achieve that, we investigated the role of ACD and autophagy in patterns of mitochondria inheritance. The novel MitoSnap model was used to allow specific tracking of old versus young organelles. We used imaging analysis of mitotic CD8+ T cells, flow cytometry readouts that allow single cell resolution, metabolic analysis and unbiased OMICS approaches to measure differences in phenotype and function between MitoSnaplo and MitoSnaphi progenies. We used adoptive cell transfers of TCR-transgenic OT-I MitoSnap cells coupled to Listeria monocytogenes-OVA infections as a tool to assess immune responses and the impact of old organelle inheritance in vivo. All conclusions rely on at least two experiments. No randomization or blinding was used. No statistical methods were used to predetermine sample sizes but our sample sizes are similar to those reported in previous publications from the field, including our own.
Animal models
All animal work was reviewed and approved by Oxford Ethical committee and the UK Home office under the project licences PPL30/3388 and P01275425. Mice were bred under specific pathogen-free conditions in-house, housed on a 12 h dark–light cycle, with a 30 min period of dawn and dusk and fed ad libitum. The temperature was kept between 20 °C and 24 °C, with a humidity level of 45–65%. Housing cages were individually ventilated and provided an enriched environment for the animals. Omp25-SnapTagfl/fl mice were kindly provided by the lab of Professor Pekka Katajisto. This strain was then bred with CD45.1 Atg16l1fl/fl Ert2Cre OT-I mice expressing a TCR specific for OVA257–264 SIINFEKL peptide41 and maintained as CD45.1 or CD45.1/2 mice. Host mice in adoptive transfer experiments were either B6.SJL.CD45.1 or C57BL/6 naive mice. The 6–16-week-old mice were considered young, and >100-week-old mice were considered aged.
CD8+ T cell isolation and activation
Spleen and inguinal lymph nodes were collected. Single-cell suspensions were used for naive CD8+ T cell isolation using EasySep Mouse Naive CD8+ T Cell Isolation Kit (Stemcell Technologies) following the manufacturer’s instructions. Purified populations were cultured (at 37 °C, 5% CO2) in T cell medium: RPMI-1640 containing HEPES buffer and L-glutamine (R5158, Sigma-Aldrich) supplemented with 10% filtered foetal bovine serum (Sigma-Aldrich), 1× penicillin–streptomycin (Sigma-Aldrich), 1× non-essential amino acids (Gibco), 50 μM β-mercaptoethanol (Gibco) and 1 mM sodium pyruvate (Gibco). T cell activation was done on anti-CD3 (5 μg ml−1) (145-2C11, BioLegend), anti-CD28 (5 μg ml−1) (37.51, BioLegend) and recombinant human or murine Fc-ICAM-1 (10 μg ml−1) (R&D Systems) coated plates. The 36–40 h post activation cells were used in downstream assays. Autophagy deletion and/or SnapTag expression were induced by culturing cells in presence of 500 nM (Z)-4-hydroxytamoxifen (Sigma-Aldrich, H7904-5MG). To determine cell division events, cells were stained with Cell Trace Violet (CTV) (Life Technologies) following manufacture’s guidelines.
SnapTag labelling protocol
MitoSnap CD8+ T cells were labelled in one or three steps. Labelling of preM/old organelles was done by collecting CD8+ T cells 12–16 h post activation and washing them in PBS (500g). The cells were incubated in T cell medium containing the first fluorescent SnapSubstrate for 30 min at 37 °C, washed in PBS and put back in culture in their original wells for further 20–24 h. When postM/young organelle labelling was also performed, cells were collected, washed and incubated with T cell medium containing 5 μM of unlabelled SnapSubstrate (Snap-Cell Block S9106S, New England Biolabs (NEB)) for 30 min at 37 °C. After washing, cells rested for 30–60 min in T cell medium and then were incubated with the second fluorescent SnapSubstrate for 30 min at 37 °C. Fluorescent cell-permeable Snap-Cell substrates (NEB) were used in the following concentrations: 3 μM (Snap-Cell 647-SiR S9102S), 3 μM (Snap-Cell TMR-Star S9105S) and 5 μM (Snap-Cell Oregon Green S9104S). BafA (20 nM) or CCCP (1 µM) were used to assess the role of autophagy and mitophagy, respectively, in the loss of OMP25 SnapTag labelling.
Cell survival and proliferation assays
Following activation, isolated MitoSnap CD8+ T cells (WT versus ATG16L1-deficient or MitoSnaphi versus MitoSnaplo first-daughter cells) were cultured in T cell medium supplemented with murine IL-2, IL-7 and IL-15 (5 ng ml−1). Cell proliferation was evaluated 3 days later, and cell survival was assessed 7 days later. For experiments evaluating the role of serine and glycine depletion, RPMI-1640 was substituted by 1× RPMI-1640 media without glucose, glycine and serine (Teknova), supplemented with 2 g l−1 of D-glucose (standard concentration found in standard RPMI medium). PHGDH inhibition was done adding 10 µM of PKUMDL-WQ-2201 to the medium.
Adoptive transfer and immunization
Fluorescence-activated cell sorting (FACS)-purified CTV-labelled MitoSnaphi or MitoSnaplo cells (equal numbers in the same experiment to allow comparison between the two groups) were intravenously injected into naive recipients. In the following day or >30 days later, mice were infected with 2 × 103 colony forming units of Listeria monocytogenes expressing ovalbumin (LM-OVA) intravenously. LM-OVA was kindly provided by Prof. Audrey Gerard (Kennedy Institute of Rheumatology, University of Oxford). LM-OVA growth was done from frozen aliquots in Brain Heart Infusion (BHI) broth (Sigma, #53286-100G). Bacteria were used for infections when reaching exponential growth. Immune responses were tracked in the blood and at the memory-phase spleens were collected.
Immunofluorescence staining and confocal microscopy
At different timepoints post stimulation (preM or mitotic/postmitotic), CD8+ T cells were collected. In some experiments, cells were incubated with 1–2 μM MitoSOX Mitochondrial Superoxide Indicator (Invitrogen) for 15 min at 37 °C before collection. The cells were washed in PBS and transferred on poly-L-lysine (Sigma-Aldrich) treated coverslips, followed by incubation for 45–60 min at 37 °C. Attached cells were fixed with 2% methanol-free paraformaldehyde in PBS (ThermoScientific) for 10 min, permeabilized with 0.3% Triton X-100 (Sigma-Aldrich) for 10 min and blocked in PBS containing 2% bovine serum albumin (Sigma-Aldrich) and 0.01% Tween 20 (Sigma-Aldrich) for 1 h at room temperature. The following antibodies were used to perform immunofluorescence stainings in murine cells: mouse anti-β-tubulin (Sigma-Aldrich), anti-mouse IgG AF488 (Abcam), anti-CD8 APC 53-6.7 (BioLegend) and anti-LC3B (D11) XP Rabbit mAb PE (Cell Signalling). DAPI (Sigma-Aldrich) was used to detect DNA. ProLong Gold Antifade Mountant (ThermoScientific) was used as mounting medium. Mitotic cells (late anaphase to cytokinesis) were identified by nuclear morphology and/or presence of two microtubule organizing centres and a clear tubulin bridge between two daughter cells. A total of 40–80 Z-stacks (0.13 μM) were acquired with a ZEISS 980 Airyscan 2 with a C-Apochromat 63×/1.2 W Corr magnification objective and the ZenBlue software. The data were analysed using Fiji/ImageJ. Thresholds for quantification were set up individually for each fluorophore. Asymmetry rates were calculated based on the integrated density (volume and fluorescence intensity measurements were considered) of cell cargoes inherited by each daughter cell. This was done by using the following calculation: (P1 − P2)/(P1 + P2), where P1 is the daughter cell with higher integrated density of CD8 or old mitochondria. Any values above 0.2 or below −0.2 were considered asymmetric, which corresponds to one daughter cell inheriting at least 1.5× more of a cell cargo than its sibling.
Planar SLBs
Planar SLBs were made as described previously66. In brief, glass coverslips were plasma-cleaned and assembled into disposable six-channel chambers (Ibidi). SLBs were formed by incubation of each channel with small unilammellar vesicles containing 12.5 mol% 1,2-dioleoyl-sn-glycero-3-[(N-(5-amino-1-carboxypentyl) iminodiacetic acid) succinyl] (nickel salt) and 0.05 mol% 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-(cap biotinyl) (sodium salt) in 1,2-dioleoyl-sn-glycero-3-phosphocholine at total phospholipid concentration 0.4 mM. Chambers were filled with human serum albumin-supplemented HEPES-buffered saline (HBS), subsequently referred to as HEPES-buffered saline/human serum albumin. Following blocking with 5% casein in PBS containing 100 μM NiSO4, to saturate NTA sites, fluorescently labelled streptavidin was then coupled to biotin head groups. Biotinylated 2C11-Fab fragments (30 molecules per square micrometre) and His-tagged ICAM-1 (200 molecules per square micrometre) and CD80 (100 molecules per square micrometre) were then incubated with the bilayers at concentrations to achieve the indicated site densities. Bilayers were continuous liquid disordered phase as determined by fluorescence recovery after photobleaching with a 10 µm bleach spot on an FV1200 confocal microscope (Olympus).
T cell immunological synapse formation on planar SLBs
Naive murine CD8+ T cells were incubated at 37 °C on an SLB. After 10 min, the cells were fixed with 4% methanol-free formaldehyde in PHEM buffer (10 mM EGTA, 2 mM MgCl2, 60 mM PIPES, 25 mM HEPES, pH 7.0) and permeabilized with 0.1% Triton X-100 for 20 min at room temperature. Anti-CD3 staining was used to identify TCR regions and actin was labelled with fluorescent phalloidin.
Total internal reflection fluorescence microscopy
Total internal reflection fluorescence microscopy was performed on an Olympus IX83 inverted microscope equipped with a four-line (405 nm, 488 nm, 561 nm and 640 nm laser) illumination system. The system was fitted with an Olympus UApON 150 × 1.45 numerical aperture objective and a Photometrics Evolve Delta EMCCD camera to provide Nyquist sampling. A quantification of fluorescence intensity was performed with ImageJ.
Flow cytometry
Blood samples used for kinetics analysis were obtained from the tail vein at weeks 1, 2 and 3 post-LM-OVA challenge. At end timepoints, spleens were collected and single-cell splenocytes were prepared by meshing whole spleens through 70 µm strainers using a 1 ml syringe plunger. When cytokine production was assessed, splenocytes were incubated at 37 °C for 1 h with 1 μM of SIINFEKL peptide, followed by 4 h in presence of SIINFEKL peptide + 10 µg ml−1 of brefeldin A (Sigma-Aldrich). Specific CD8+ T cells were evaluated by incubation with SIINFEKL257-264-APC-labelled or SIINFEKL257-264-BV421-labelled tetramers (NIH Tetramer Core Facility at Emory University). Erythrocytes were lysed by red blood cell lysis buffer (Invitrogen). Conjugated antibodies used for surface staining were: anti-CD127 A7R34, anti-CD25 PC61 (AF700, PE-Cy7, BioLegend; APC, eBioscience), anti-CD44 IM7 (AF700, BV785, PE, PerCPCy5.5, BioLegend), anti-CD45.1 A20 (BV785, FITC, PB, BioLegend), anti-CD45.2 104 (AF700, BV711, FITC, BioLegend), anti-CD62L MEL-14 (FITC, BioLegend; eF450, eBioscience), anti-KLRG1 2F1 (BV711, BV785, BioLegend), anti-CD8 53-6.7 (BV510, BV605, FITC, PE, BioLegend) and anti-TCRβ H57-597 (APC-Cy7, PerCPCy5.5, BioLegend). Cells were incubated for 20 min at 4 °C. When intracellular staining was performed, cells were fixed/permed with 2× FACS Lysis Solution (BD Biosciences) with 0.08% Tween 20 (Sigma-Aldrich) for 10 min at room temperature, washed in PBS and incubated for 1 h at RT with anti-IL-2 JES6-5H4 (APC, BioLegend), anti-IFN-γ XMG1.2 (BV421, BioLegend) and anti-TNF MP6-XT22 (PE-Cy7, Thermo Fisher). Identification of viable cells was done by fixable near-IR dead cell staining (Life Technologies). TMRM (100 nM) and MitoSOX (2 µM) labelling was done before staining of surface markers, by incubating cells for 15 min at 37 °C in T cell medium. All samples were washed and stored in PBS containing 2% FBS (Sigma-Aldrich) and 5 mM of EDTA (Sigma-Aldrich) before acquisition. Stained samples were acquired on a FACS LSR II (R/B/V) or a Fortessa X-20 (R/B/V/YG) flow cytometer (BD Biosciences) with FACSDiva software. Data analysis was done using FlowJo software (FlowJo Enterprise, version 10.10, BD Biosciences).
Cell sorting (FACS)
After activation, CTV- and SnapSubstrate-labelled MitoSnap CD8+ T cells were collected and stained for phenotypical markers (anti-CD44 IM7, anti-CD45.1 A20, anti-CD45.2 104 and anti-CD8 53-6.7 conjugated to different fluorophores depending on experiment, all BioLegend). Dead cells were excluded by staining cells with a fixable Live/Dead dye (Invitrogen, L34993 or L34957). Subpopulations of interest were sorted on a FACS Aria III cell sorter (BD Biosciences).
Metabolic reliance measured by protein translation (Scenith)
We used a modified version of the Scenith assay45, which describes a high correlation between protein translation and ATP production. New protein synthesis was measured using the Click-iT Plus OPP Protein Synthesis Assay (Thermo Fisher, C10456/C10457) according to manufacturer’s protocol. In short, the cells were incubated in T cell medium for 30 min at 37 °C without any metabolic inhibitors or in presence of 1 μM oligomycin (Merck), 100 mM 2DG (Merck), a combination of both, 1 μM of SHIN1 (Cambridge Bioscience), 20 μM of PKUMDL-WQ-2201 (Sigma-Aldrich) or 1 μM of LY345899 (Sigma-Aldrich). This was followed by incubation with 10 μM of alkynylated puromycin analogue OPP for 30 min at 37 °C. Click Chemistry was performed with Alexa Fluor 488 or Alexa Fluor 594 dye picolyl azide. Metabolic reliance was assessed by comparing the OPP gMFI, used as an indicator of the relative translation rate, of inhibited samples to the vehicle control.
Western blot
Following (Z)-4-hydroxytamoxifen (Sigma-Aldrich, H7904-5MG) treatment for 24 h and/or BafA treatment (10 nM) for 2 h or not, cells were washed with PBS and lysed in RIPA lysis buffer (Sigma-Aldrich) supplemented with complete Protease Inhibitor Cocktail (Roche) and PhosSTOP (Roche). Protein concentration was calculated by using the BCA assay (Thermo Fisher). Samples were diluted in 4× Laemmli sample buffer (Bio-Rad) and boiled at 100 °C for 5 min. A total of 20 µg protein per sample were used for SDS–polyacrylamide gel electrophoresis analysis. NuPAGE Novex 4–12% Bis–Tris gradient gel (Invitrogen) with MOPS running buffer (Invitrogen) was used. Proteins were transferred to a PVDF membrane (Merck Millipore) and blocked with 5% skimmed milk-TBST (TBS 10× (Sigma-Aldrich) diluted to 1× in distilled water containing 0.1% Tween 20 (Sigma-Aldrich)) for 1 h. Membranes were incubated at 4 °C overnight with primary antibodies diluted in 1% skimmed milk-TBST and at room temperature for 1 h with secondary antibodies diluted in 1% skimmed milk-TBST supplemented 0.01% SDS. Primary antibodies used were: anti-ATG16L1, clone EPR15638 (Abcam, ab187671) and anti-GAPDH and clone 6C5 (Sigma-Aldrich, MAB374). Secondary antibodies used were: IRDye 680LT Goat anti-Mouse IgG (H+L) (Licor, 926-680-70) and IRDye 800CW Goat anti-Rabbit IgG (H+L) (Licor, 926-322-11). Images were acquired using the Odyssey CLx Imaging System. The data were analysed using Image Studio Lite or Fiji.
Mitochondrial isolation and flow cytometry (MitoFlow)
Autophagy-sufficient (Atg16l1fl/+ Omp25fl/+ Ert2Cre) and autophagy-deficient (Atg16l1fl/fl Omp25fl/+ Ert2Cre) MitoSnap CD8+ T cells were activated, labelled for preM/old (SNAP-Cell TMR-Star or 647-SiR, NEB) and postM/young organelles (SNAP-Cell Oregon Green, NEB), as previously described, and after 40 h washed with complete T cell medium. Cell pellets were resuspended in ice-cold mitochondria isolation buffer (320 mM sucrose, 2 mM EGTA,10 mM Tris–HCl, at pH 7.2 in water) and homogenized with a Dounce homogenizer with a 2 ml reservoir capacity (Abcam). We performed 20 strokes with a type B pretzel. The homogenizer was rinsed with distilled water before each sample was processed to avoid cross-contamination. Differential centrifugation of homogenates was done at 1,000g (4 °C for 8 min), which resulted in a pellet containing whole cells and isolated nuclei first. The supernatant containing the mitochondria was then transferred into new tubes and centrifuged at 17,000g (4 °C for 15 min). Enriched mitochondria, which appeared as brown-coloured pellets, were fixed in 1% paraformaldehyde in 0.5 ml PBS on ice for 15 min, followed by a wash with PBS. Mitochondria were resuspended in blocking buffer containing anti-Tom20-BV421 antibody for 20 min at room temperature. After washing with PBS, mitochondria were resuspended in 250 μl filtered (0.2 μm) PBS and acquired using a BD Fortessa X-20 flow cytometer. The threshold for SSC-A (log-scale) was set to the minimum value (20,000) to allow acquisition of subcellular particles. Submicron particle size reference beads (0.5, 1 and 2 µm, Thermo Fisher Scientific) were also used to identify mitochondria.
Seahorse assay
MitoSnaphi and MitoSnaplo cells were purified by FACS and their OCR were measured using a XF96 MitoStress Test (Seahorse Agilent, 103015-100). Activated CD8+ T cells were washed in RPMI-1640 without sodium bicarbonate, 10 mM glucose, 1% FCS, 2 mM pyruvate and seeded in a XF plate (Agilent, 103793-100) coated with poly-L-lysine (Sigma-Aldrich) at equal densities in corresponding assay medium (XF Assay Medium, 103680-100) pH 7.4 supplemented with 10 mM glucose, 1 mM sodium pyruvate and 2 mM L-glutamine. Test compounds were sequentially injected to obtain the following concentrations: 1 µM oligomycin, 1.5 µM FCCP, 1 µM rotenone and 1 µM antimycin A. OCRs were normalized to cell number using CyQuant (Molecular Probes).
ATP synthesis assay
Sorted MitoSnaphi and MitoSnaplo CD8+ T cells were boiled in 100 mM Tris, 4 mM EDTA, pH 7.74 buffer for 2 min at 100 °C. Following centrifugation, the supernatant was used for analysis. ATP levels were assessed using the ATP Bioluminescence Assay Kit CLS II (Roche) following the manufacturer’s instructions. The samples and ATP standard mixtures were swiftly combined with an equal volume of luciferase and promptly measured in a luminometer (BMG CLARIOstar Plus microplate reader). Normalization was performed by adjusting values based on the total number of sorted cells. The experiment was performed twice. Each experiment was done with two samples per group (each sample pooled from two biological replicates).
Proteomics
Proteomics analysis was done as previously described67. CD8hi and CD8lo or MitoSnaphi and MitoSnaplo daughter cells following naive CD8+ T cell activation were purified by FACS. The cell pellets were washed 2× in PBS before being stored at −80 °C before proteomics analysis. Samples were resuspended in 200 µl of S-Trap lysis buffer (10% SDS, 100 mM triethylammonium bicarbonate) and sonicated for 15 min (30 s on, 30 s off, 100% amplitude, 70% pulse). The samples were centrifuged and supernatants were transferred to fresh tubes. Protein quantification was done using the Micro BCA Protein Assay Kit (Thermo Fisher). A total of 150 μg of protein was processed using S-Trap mini columns (Protifi, #CO2-mini-80). The samples were digested overnight with 3.75 μg of trypsin (Thermo Fisher, Pierce Trypsin Protease MS-Grade, #90057) with a second digest with the same amount of trypsin for 6 h the following day. The peptides were extracted, dried under vacuum and resuspended to 50 μl with 1% formic acid (Thermo Fisher, #85178) and quantified using the Pierce Quantitative Fluorometric Peptide Assay (Thermo Fisher, #23290).
Peptides were injected onto a nanoscale C18 reverse-phase chromatography system (UltiMate 3000 RSLC nano, Thermo Fisher) and electrosprayed into an Orbitrap Exploris 480 mass spectrometer (MS) (Thermo Fisher). For liquid chromatography the following buffers were used: buffer A (0.1% formic acid in Milli-Q water (v/v)) and buffer B (80% acetonitrile and 0.1% formic acid in Milli-Q water (v/v)). The samples were loaded at 10 μl min−1 onto a trap column (100 μm × 2 cm, PepMap nanoViper C18 column, 5 μm, 100 Å, Thermo Fisher) equilibrated in 0.1% trifluoroacetic acid. The trap column was washed for 3 min at the same flow rate with 0.1% trifluoroacetic acid then switched in-line with a Thermo Fisher, resolving C18 column (75 μm × 50 cm, PepMap RSLC C18 column, 2 μm, 100 Å). Peptides were eluted from the column at a constant flow rate of 300 nl min−1 with a linear gradient from 3% buffer B to 6% buffer B in 5 min, then from 6% buffer B to 35% buffer B in 115 min, and finally from 35% buffer B to 80% buffer B within 7 min. The column was then washed with 80% buffer B for 4 min. Two blanks were run between each sample to reduce carry-over. The column was kept at a constant temperature of 50 °C. The data was acquired using an easy spray source operated in positive mode with spray voltage at 2.60 kV, and the ion transfer tube temperature at 250 °C. The MS was operated in DIA mode. A scan cycle comprised a full MS scan (m/zrange from 350 to 1,650), with RF lens at 40%, AGC target set to custom, normalized AGC target at 300%, maximum injection time mode set to custom, maximum injection time at 20 ms, microscan set to 1 and source fragmentation disabled. MS survey scan was followed by MS/MS DIA scan events using the following parameters: multiplex ions set to false, collision energy mode set to stepped, collision energy type set to normalized, HCD collision energies set to 25.5%, 27% and 30%, orbitrap resolution 30000, first mass 200, RF lens 40%, AGC target set to custom, normalized AGC target 3,000%, microscan set to 1 and maximum injection time 55 ms. The data for both MS scan and MS/MS DIA scan events were acquired in profile mode.
Analysis of the DIA data was carried out using Spectronaut Biognosys, AG (version 14.7.201007.47784 for CD8hi and CD8lo cells obtained from young, Atg16l1-deficient and old mice; version 17.6.230428.55965 for MitoSnaphi and MitoSnaplo cells). The data were analysed using the direct DIA workflow, with the following settings: imputation, profiling and cross run normalization were disabled; data Filtering to Qvalue; Precursor Qvalue Cutoff and Protein Qvalue Cutoff (Experimental) set to 0.01; maximum of two missed trypsin cleavages; PSM, protein and peptide false-discovery rate levels set to 0.01; cysteine carbamidomethylation set as fixed modification and acetyl (N-term), deamidation (asparagine, glutamine), oxidation of methionine set as variable modifications. The database used for CD8hi and CD8lo cells was mouse_swissprot_isoforms_extra_trembl_06_20.fasta (2020-06) and for mitosnap samples was the Mus musculus proteome obtained from uniprot.org (2022-02). Data filtering, protein copy number and concentration quantification was performed in the Perseus software package, version 1.6.6.0. Copy numbers were calculated using the proteomic ruler as described33. The samples were grouped according to the condition. Outliers were excluded following analysis of histone counts in each sample. The P values were calculated via a two-tailed, unequal-variance t-test on log-normalized data. Elements with P values <0.05 were considered significant, with a fold-change cut-off >1.5 or <0.67.
Single cell transcriptomics
Single cell RNA sequencing libraries were prepared using the Chromium Single Cell 3′ GEX v3.1 assay (10x Genomics). In short, cell suspensions were encapsulated into gel beads in emulsion using the Chromium Controller. Within each gel bead in emulsion, cell lysis and barcoded reverse transcription of RNA occurred, followed by complementary DNA amplification. The amplified cDNA underwent library construction via fragmentation, end-repair, A-tailing, adaptor ligation and index PCR. Final libraries were sequenced on an Illumina NovaSeq 6000 system. Initial data processing was conducted with Cell Ranger 7.2.0.
Filtered output matrices were processed using Seurat. After loading the data and assigning unique identifiers to each dataset, cells with more than 10% mitochondrial gene content were excluded to ensure data quality. The datasets were normalized using SCTransform, and PCA was conducted for dimensionality reduction. Integration of the datasets was achieved using the Harmony algorithm, followed by clustering and differential expression analysis. Finally, the integrated data were visualized using UMAP. UMAP projections of resulting clusters and gene expression of gene sets were generated using Loupe Browser. This methodology enabled a robust analysis while accounting for technical variations and maintaining biological integrity.
Trajectory analysis and visualization
To investigate the pseudotemporal dynamics of our single-cell RNA sequencing data, we utilized Monocle368, an advanced tool designed for trajectory inference and differential expression analysis. In previous steps, using the Seurat package, the cells were preprocessed, and size factors were estimated and normalized. Dimensionality reduction was performed using UMAP to capture the intrinsic structure of the data, followed by clustering to identify distinct cell populations. These clusters were annotated based on known marker genes, and the cluster annotations were integrated into the metadata for further analysis.
For trajectory inference, we converted the Seurat object to a Monocle3 object69. Separate trajectory analyses were conducted for the MitoSnaphi and MitoSnaplo original identities. Each subset of cells was reclustered using UMAP, and principal graphs were constructed to capture the developmental trajectories of the cells. Pseudotime values were then computed to represent the progression of cells along these inferred trajectories.
Mitochondrial architecture analysis
MitoSnaphi and MitoSnaplo cells were FACS isolated and prepared for immunofluorescence as described above. Z-stacks (0.13 μm) were acquired with a ZEISS 980 Airyscan 2 with a C-Apochromat 63×/1.2 W Corr magnification objective and the ZenBlue software. Image voxels were made isotropic via upscaling using the ‘zoom’ function of the SciPy Multidimensional image package70. A feathering algorithm, to pick up on mitochondria outlines, in combination with multi-Otsu thresholding and a custom set of operators (that is, erosions, closing and filling of holes) from scikit-image (Python) were used for segmentation of individual mitochondria71. Both the Tom20 and MitoSnap channels were combined to enable identification of all mitochondria. Segments smaller than 500 pixels (noise) were excluded from further analysis. Separate organelles were given a unique identifier using the ‘label’ function from scikit-image. Each segmented organelle was used to create a mask on the isotropic pixel converted images, which allowed quantification of fluorescence intensity values (calculated as the average pixel intensity/number of pixels in the segmented volume). Segmented organelles had their geometric parameter measurements extracted and standardized, and included sphericity, roundness, compactness, surface area72 and volume. Mitochondrial complexity was calculated as previously described73.
The data were visualized with UMAP. Only geometric parameters were used to remove any bias in the analysis. A Gaussian kernal density estimate was plotted to produce the contour/density map of UMAP points for visualization of mitochondria distribution shifts. Mitochondria were grouped and analysed for geometric parameters based on their cell of origin (MitoSnaphi versus MitoSnaplo) and intensity of SnapTag labelling. MitoSnaplo cells exhibited only mitochondria with low intensity of SnapTag labelling. In MitoSnaphi cells, low intensity was determined by intensity values below the maximum intensity value found in MitoSnaplo cells; mid intensity was determined as values above this threshold, but below the 75th percentile value found in MitoSnaphi cells; high mitochondria were identified using SnapTag labelling intensity values above the 75th percentile value for found in MitoSnaphi cells.
Metabolomics
CTV-labelled naive MitoSnap CD8+ T cells (WT-Atg16l1fl/+ Omp25fl/+ Ert2Cre or knockout (KO)-Atg16l1fl/fl Omp25fl/+ Ert2Cre) were activated on anti-CD3, anti-CD28 and Fc-ICAM-1 coated plates for 40 h. The cells were cultured in T cell medium containing 500 nM Z-4-hydroxytamoxifen (4OHT). Then, 16 h post activation, the cells were collected and labelled with Snap-Cell 647-SiR (preM/old mitochondria) and cultured for further 24 h in presence of 13C-glucose (2 g l−1) or 13C-serine (30 mg l−1). Metabolites were extracted from cell pellets of sorted MitoSnaphi and MitoSnaplo daughter cells. A total of 10 µl of each sample was loaded into a Dionex UltiMate 3000 LC System (Thermo Scientific Bremen, Germany) equipped with a C18 column (Acquity UPLC -HSS T3 1.8 µm; 2.1 × 150 mm, Waters) coupled to a Q Exactive Orbitrap MS (Thermo Scientific) operating in negative ion mode. A step gradient was carried out using solvent A (10 mM TBA and 15 mM acetic acid) and solvent B (100% methanol). The gradient started with 0% of solvent B and 100% solvent A and remained at 0% B until 2 min post injection. A linear gradient to 32.3% B was carried out until 7 min and increased to 36.3% until 14 min. Between 14 and 26 min, the gradient increased to 90.9% of B and remained at 90.9% B for 4 min. At 30 min the gradient returned to 0% B. The chromatography was stopped at 40 min. The flow was kept constant at 0.25 ml min−1 and the column was placed at 40 °C throughout the analysis. The MS operated in full scan mode (m/z range 70.0000–1,050.0000) using a spray voltage of 4.80 kV, capillary temperature of 300 °C, sheath gas at 40.0 and auxiliary gas at 10.0. The AGC target was set at 3 × 106 using a resolution of 70,000, with a maximum IT fill time of 512 ms. Data collection was performed using the Xcalibur software (Thermo Scientific). The data analyses were performed by integrating the peak areas (El-Maven–Polly, Elucidata).
Statistical analysis
To test if data point values were in a Gaussian distribution, a normality test was performed before applying parametric or non-parametric statistical analysis. When two groups were compared, paired or unpaired Student’s t-test or Mann–Whitney test were applied. When comparisons were done across more than two experimental groups, analyses were performed using a Dunn test with Bonferroni correction or one-way analysis of variance (ANOVA) or two-way ANOVA with a post hoc Tukey’s test multiple testing correction. The P values were considered significant when <0.05, and the exact P values are provided in the figures. All analyses were done using GraphPad Prism 10 software.
Inclusion and ethics statement
We affirm our commitment to ethical integrity, fairness, and inclusion by engaging diverse collaborators and local researchers throughout the study, while carefully and transparently describing author contributions.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.