Abstract

Background

   Apoptosis, a form of programmed cell death, is critical for the
   development and homeostasis of the immune system. Chimeric antigen
   receptor T (CAR-T) cell therapy, approved for hematologic cancers,
   retains several limitations and challenges associated with ex vivo
   manipulation, including CAR T-cell susceptibility to apoptosis.
   Therefore, strategies to improve T-cell survival and persistence are
   required. Mesenchymal stem/stromal cells (MSCs) exhibit
   immunoregulatory and tissue-restoring potential. We have previously
   shown that the transfer of umbilical cord MSC (UC-MSC)-derived
   mitochondrial (MitoT) prompts the genetic reprogramming of CD3^+ T
   cells towards a T[reg] cell lineage. The potency of T cells plays an
   important role in effective immunotherapy, underscoring the need for
   improving their metabolic fitness. In the present work, we evaluate the
   effect of MitoT on apoptotis of native T lymphocytes and engineered
   CAR-T cells.

Methods

   We used a cell-free approach using artificial MitoT (Mitoception) of
   UC-MSC derived MT to peripheral blood mononuclear cells (PBMCs)
   followed by RNA-seq analysis of CD3^+ MitoT^pos and MitoT^neg sorted
   cells. Target cell apoptosis was induced with Staurosporine (STS), and
   cell viability was evaluated with Annexin V/7AAD and TUNEL assays.
   Changes in apoptotic regulators were assessed by flow cytometry,
   western blot, and qRT-PCR. The effect of MitoT on 19BBz CAR T-cell
   apoptosis in response to electroporation with a non-viral
   transposon-based vector was assessed with Annexin V/7AAD.

Results

   Gene expression related to apoptosis, cell death and/or responses to
   different stimuli was modified in CD3^+ T cells after Mitoception.
   CD3^+MitoT^pos cells were resistant to STS-induced apoptosis compared
   to MitoT^neg cells, showing a decreased percentage in apoptotic T cells
   as well as in TUNEL^+ cells. Additionally, MitoT prevented the
   STS-induced collapse of the mitochondrial membrane potential (MMP)
   levels, decreased caspase-3 cleavage, increased BCL2 transcript levels
   and BCL-2-related BARD1 expression in FACS-sorted CD3^+ T cells.
   Furthermore, UC-MSC-derived MitoT reduced both early and late apoptosis
   in CAR-T cells following electroporation, and exhibited an increasing
   trend in cytotoxic activity levels.

Conclusions

   Artificial MitoT prevents STS-induced apoptosis of human CD3^+ T cells
   by interfering with the caspase pathway. Furthermore, we observed that
   MitoT confers protection to apoptosis induced by electroporation in
   MitoT^pos CAR T-engineered cells, potentially improving their metabolic
   fitness and resistance to environmental stress. These results widen the
   physiological perspective of organelle-based therapies in immune
   conditions while offering potential avenues to enhance CAR-T treatment
   outcomes where their viability is compromised.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12967-024-05627-4.

   Keywords: Mesenchymal stromal/stem cells, Chimeric antigen receptor T
   (CAR-T) cells, Mitochondria transfer, Induced-apoptosis

Background/introduction

   Apoptosis is an evolutionary conserved form of controlled cell death
   [[40]1]. Either defective or excessive apoptosis is a main contributor
   to cancer, autoimmunity, or immune deficiency [[41]2]. In the immune
   system, apoptosis is critical for the development of T and B
   lymphocytes, ensuring the elimination of self-reactive cells and those
   unable to recognize the major histocompatibility complex (MHC) before
   they enter into circulation [[42]3, [43]4]. Apoptosis also plays a role
   in activation-induced cell death (AICD) after lymphocyte activation by
   antigen recognition since it is essential in the contraction phase of
   the immune response, eliminating activated lymphocytes while preserving
   the long-lived memory cell population [[44]5]. Different physiological
   and pathological immune processes stem from increased apoptosis within
   specific immune cell subsets. Such is the case of acquired immune
   deficiency syndrome (AIDS), where CD4^+ T lymphocytes die due to the
   infection with the human immunodeficiency virus (HIV-1) or the
   increased predisposition to undergo apoptosis of CD4^+ and CD8^+
   subsets in old people as a consequence of immunosenescence [[45]6].

   On the other hand, CAR-T cells have become of great interest for
   treating hematological tumors. Recent observations also highlight the
   potential of CAR-T therapy in other applications including solid
   tumors, chronic infections, cardiac fibrosis, autoimmune diseases, and
   cellular senescence [[46]7]. Nevertheless, CAR T-cell therapy retains
   several limitations, like poor trafficking and tumor infiltration,
   limited persistence, life-threatening cell-associated toxicities, in
   addition to the ex vivo manipulation that increases CAR T-cell
   senescence and susceptibility to apoptosis [[47]8, [48]9]. CAR-T cells
   approved or in clinical trials require a viral transfection to provide
   long-term CAR expression, that holds high production costs and can take
   up to 15 days to generate the infusion product, while the nonviral
   transposon-based vector can generate CAR-T cells in 8 days, showing
   long-term persistence and anti-tumor response in animal models.
   However, electroporation with transposon plasmids decreases PBMCs
   viability to ~ 20% [[49]10, [50]11]. Therefore, new therapeutic
   strategies to overcome these challenges focused on improving T-cell
   survival are required.

   Mesenchymal stem/stromal cells (MSCs), characterized by multipotent and
   self-renewal capabilities [[51]12], exert anti-inflammatory and
   immunomodulatory effects through distinct mechanisms such as paracrine
   signals, including the secretion of soluble factors and extracellular
   vesicles [[52]13]. MSCs can also exert therapeutic effects by the
   transfer of functional mitochondria (MitoT) to damaged cells by
   restoring metabolic and energetic needs [[53]14–[54]16]. We have
   recently demonstrated that MitoT from umbilical cord (UC)-MSC targets a
   range of immunocompetent cells, leading to marked changes in immune
   function. Indeed, MitoT to T CD4^+ cells induced a switch toward a
   functional regulatory T (T[reg]) cell phenotype that curbed the
   progression of Graft-versus-host disease (GVHD) [[55]17].

   Although MitoT is known to rescue damaged cells from several diseases
   by enhancing their anti-apoptosis ability [[56]18, [57]19], the effect
   of MSC-derived MitoT on the survival of apoptotic-challenged native
   lymphocytes and CAR-T cells has not been addressed. We hypothesized
   that MitoT could target T-cell apoptosis, regulating cell survival and
   persistence, two key factors in translational cell therapies.

   Metabolic pathways not only provide essential building blocks for
   unmanipulated T and CAR-T cells during their differentiation and
   expansion but also significantly influence the outcomes of T-based cell
   therapy, as recent studies have indicated [[58]20, [59]21].

   To discriminate the level of mitochondria (MT)-dependent
   immunoregulatory impact without the interference of other mechanisms,
   here, we used mitoception, which is a cell-free approach using the
   artificial transfer of isolated MT. Hence, purified UC-MCS-derived MT
   are incubated with T cells and assayed for MitoT and ensuing functional
   and metabolic changes. Herein we present evidence supporting the
   hypothesis that the delivery of UC-MSC-derived MT to human T
   lymphocytes through the artificial transfer of MT enhances cell
   survival of native and engineered CAR-T cells (Fig. [60]1).

Fig. 1.

   [61]Fig. 1
   [62]Open in a new tab

   Experimental workflow of MitoT to native and CAR engineered T cells.
   This schematic diagram illustrates the experimental methodology used to
   isolate mitochondria from umbilical cord MSCs after their transfer into
   T cells and subsequently challenge them by staurorsporine or through
   DNA electroporation for CAR-T generation. Apoptosis was then assessed
   using various markers and membrane potential measurements

Methods

Transcriptome profiling and gene expression analysis.

   Whole-genome RNA-Sequencing (RNA-Seq) datasets were downloaded from the
   Gene Expression Omnibus [63]GSE140483 (SRA numbers
   SRR10466068–SRR10466075) [[64]17]. Adapter sequences were removed from
   the raw reads using Trimmomatic version 0.39 [[65]22]. A five-base
   sliding window quality filtering was also performed, eliminating reads
   when the average quality of a window fell below Q < 30. Reads shorter
   than 15 nucleotides after trimming were also discarded. The resulting
   data was mapped to the human genome GRCh38 [[66]23] using HISAT2
   version 2.2.1 [[67]24]. Gene expression and differential analysis were
   performed on annotated genes in the reference genome by using the
   Transcriptome assembly and differential expression analysis for RNA-Seq
   platform Cufflinks ([68]http://cufflinks.cbcb.umd.edu/) [[69]25], using
   the GTF annotation file recovered from the Ensembl database for the
   human genome GRCh38. Genes were considered as statistically
   differentially expressed (DE) when presenting an absolute fold change
   (FC) > 1.5 and a p-value < 0.05. Only protein-coding sequences (205
   genes) were considered in the following analysis. Volcano plots were
   created with the analysis results tool CummeRbund following the
   protocol previously described [[70]25]. Graphical representation of p
   values versus fold change for the expression datasets was visualized as
   volcano plot.

Differential gene expression and pathway analysis

   We used the web-based Reactome Pathway Database tool
   ([71]https://reactome.org) and the Cytoscape v.3.9.0 software
   ([72]https://cytoscape.org) to identify pathways and gene networks that
   were enriched for DE genes between previously FACS-sorted T CD3^+
   MitoT^pos and CD3^+ MitoT^neg cells. Gene Ontology (GO) enrichment was
   done using the R package “GOfuncR” [[73]26], considering 100 random
   sets. False Discovery Rate (FDR) multiple testing corrected
   p-values < 0.05 were used for GO enrichment. Significant DE genes
   belonging to the GO terms: leukocyte mediated cytotoxicity
   (GO:0001909), regulation of leukocyte mediated cytotoxicity
   (GO:0001910), apoptotic process (GO:0006915), defense response
   (GO:0006952), response to external stimulus (GO:0009605), response to
   organic substance (GO:0010033), programmed cell death (GO:0012501),
   regulation of response to external stimulus (GO:0032101), response to
   chemical (GO:0042221), regulation of apoptotic process (GO:0042981),
   positive regulation of apoptotic process (GO:0043065), regulation of
   programmed cell death (GO:0043067), positive regulation of programmed
   cell death (GO:0043068), regulation of cell differentiation
   (GO:0045595), lymphocyte activation (GO:0046649), positive regulation
   of response to stimulus (GO:0048584), negative regulation of response
   to stimulus (GO:0048585), cellular response to chemical stimulus
   (GO:0070887), cellular response to organic substance (GO:0071310),
   cellular response to cytokine stimulus (GO:0071345), were further
   analyzed.

Cell isolation and culture

   Peripheral blood samples were collected from healthy donors with
   previous informed consent, and Ficoll-Paque Premium density gradient
   media (GE #17544203) was used for the isolation of PBMCs. For CAR-T
   experiments, PBMCs were obtained from leukocyte reduction filters
   (RS—Haemonetics) and purified using Ficoll-Paque. Following the general
   guidelines for the selection of blood donors, the inclusion criteria,
   included male and female donors, between 25 and 45 years of age, weight
   more than 50 kg, good level of general health, no anemia or high blood
   pressure, or any flu symptoms. No chronic diseases, heart disease, or
   diabetes. The PBMCs were cultured in lymphocyte media (RPMI media
   supplemented with 10% of FBS, 100 U/mL penicillin, 100 µg/mL
   streptomycin, 1 mM sodium pyruvate, 1× non-essential amino acids, 2 mM
   l-glutamine and 25 μM β-mercaptoethanol).

   The human UC-MSCs were isolated from healthy donors and characterized
   according to the International Society for Cell and Gene Therapy (ISCT)
   minimal criteria, as we previously described [[74]27]. The UC-MSCs used
   in this present study were plastic-adherent when maintained in standard
   culture conditions, they express CD73, CD90 and CD105, and lack
   expression of CD45, CD34, CD14, HLA-DR and CD19 surface molecules, and
   approved for clinical use [[75]28].

   All experiments were performed using early passage (P3–P7) UC-MSCs
   cultured in humidified incubators at 37 °C with 5% CO[2], using DMEM
   High Glucose media supplemented with 10% fetal bovine serum (FBS), 2 mM
   l-glutamine and 100 U/mL penicillin, 100 µg/mL streptomycin. All the
   procedures presented in this work were approved by the Ethics Committee
   of Universidad de los Andes, Chile (#CEC202729) and by the Ethics
   Committee of the Brazilian National Cancer Institute
   (CEP-INCA)—51589521.4.0000.5274, in conformity with the Helsinki
   Declaration.

Mitochondria isolation and artificial transfer (Mitoception)

   MT in UC-MSCs were labeled with MitoTracker Green FM, MitoTracker Deep
   Red (500 nM, Invitrogen #M7514; #[76]M22426), or MitoView720 (200 nM,
   Biotium #70068), and isolated using a commercial kit following the
   manufacturer’s recommendations (ThermoFisher Scientific #89874).
   Freshly isolated MT from 10^6 to 10^7 UC-MSCs were resuspended in
   lymphocyte media and transferred to PBMCs in a ratio of 1:10 (MT in
   quantities equivalent to UC-MSC: PBMC) following the Mitoception
   protocol [[77]29]. The MitoTracker fluorescence signal stability on
   CD3^+ cells after MitoT was evaluated. (Additional file [78]1: Figure
   S1). Immediately, cells were cultured under standard conditions, and
   after 24 h cells were washed and used for experimental procedures
   (Fig. [79]1).

T cell-electroporation

   For gene transfer, we used the nonviral transposon-based vector
   PiggyBac, coding for the 19BBz CAR. 3 × 10^7 PBMCs were Mitocepted with
   MitoView720 stained UC-MSC-derived MT in a ratio 1:10 (UC-MSC: PBMC) or
   as control cells were only cultured with lymphocyte media and were
   cultured for 24 h at 37 °C, 5% CO[2]. Subsequently, mitocepted PBMCs
   were electroporated in a sterile 0.2 cm cuvette (Mirus Biotech) with
   the Lonza Nucleofector II electroporation system under the program U-14
   with 100 µL of 1SM buffer [[80]30], 10 µg of PBCAG 19BBz (transposon)
   and 20 µg PBase (transposase), or just with PBase plasmid as control
   (mock condition). After electroporation, cells were resuspended in 1 mL
   pf RPMI medium supplemented with 2 mM l-glutamine, 50 U/mL IL-2
   (ThermoFisher Scientific #PHC0021), 100 U/mL penicillin, 100 µg/mL
   streptomycin, 10 mM HEPES, 0.1 mM pyruvate and 20% fetal calf serum and
   cultured at 37 °C, 5% CO[2]. After 2 h, these cells were activated with
   T Cell TransAct CD3/CD28 (Miltenyi Biotec) at a concentration of
   15 µL/mL. On the next day, cell viability was evaluated using Annexin
   and 7AAD (Fig. [81]1). The PBCAG GFP transposon plasmid was acquired in
   Addgene (#40973), and the 19BBz transgene was inserted into the GFP by
   cloning with restriction enzymes Agel and Notl. A Myc tag was added to
   CAR 19BBz between CD8a signal peptide and scFv to allow CAR detection
   by flow cytometry. The transposase containing plasmid was kindly
   provided by Dr. Joseph Loturco at the University of Connecticut
   [[82]31].

Cell viability by Annexin V/7AAD assay

   For cell viability and apoptosis assays, 500,000 PBMCs were seeded on a
   48-well plate and mitocepted with UC-MSC-derived MT previously labeled
   with MitoTracker Green at a 1:10 ratio. After 24 h, apoptosis was
   induced with 10 μM or 20 μM of STS (Abcam #ab120056) for 4 h, as other
   authors have previously demonstrated protection from apoptosis of human
   T lymphocytes in vitro [[83]32]. Then, cells were collected and stained
   with 1:200 anti-CD3-PE (BD #555333) or anti-CD3-PECy7 (BD #557851)
   antibodies for 30 min at 4 °C protected from light. After washing with
   PBS, cells were labeled with 1:50 Annexin V-APC (Biolegend #640919) and
   1:25 7AAD-PerCP-Cy5.5 (Biolegend #420404) using a commercial binding
   buffer (Biolegend #422201). For CAR-T experiments, PBMCs were collected
   and stained with 1:500 anti-CD3-PECy7 (Cytek, #60-0036), 1:100
   anti-c-Myc9E10-AlexaFluor647 (Santa Cruz #SC-40), 1:200 Annexin V-FITC
   (Biolegend #640945), and 1:100 7AAD-PERCP. Data were acquired using
   FACS Canto II, analyzed with the FlowJo v.10 software, and expressed
   with the Mean Fluorescence Intensity (MFI) values.

TUNEL assay

   Apoptotic cell death was analyzed using the TUNEL assay kit (Abcam,
   #ab66110) according to manufacturer protocol. Briefly, 500,000 PBMCs
   were seeded on a 24-well plate and mitocepted with UC-MSC-derived MT
   previously labeled with MitoTracker Deep Red at a 1:10 ratio. After
   24 h, apoptosis was induced for 4 h with 5 µM or 10 µM STS at 37 °C.
   Cells were then washed, stained with anti-CD3-PECy7, fixed, and labeled
   with BrdU-Red and DNA labeling solution, following the manufacturer’s
   recommendations. Finally, cells were collected and analyzed immediately
   using BD FACS Canto II, and FlowJo v.10 software.

Western blot analysis

   Mitocepted PBMCs were seeded on a 6-well plate at a confluence of
   1 × 10^7 cells per well, and 24 h after, apoptosis was induced with
   10 µM STS for 5 h. CD3^+MitoT^pos and CD3^+MitoT^neg cells were
   FACS-sorted by using BD FACS Aria Fusion and resuspended with 50 µL of
   cold Radioimmunoprecipitation Assay (RIPA) buffer containing
   protease/phosphatase inhibitors (Cell Signaling Technology, #5872).
   Samples were then sonicated for 5 min at high intensity and centrifuged
   at 16,000×g for 15 min at 4 °C to discard cellular debris. Supernatants
   were collected, and protein quantification was performed using the
   Bradford method following the manufacturer’s instructions (Bio-Rad,
   #500-006). 20–30 µg of protein were loaded and electrophoresed into 10%
   polyacrylamide gels. Next, proteins were transferred to a
   nitrocellulose membrane and were incubated with a blocking solution (5%
   Bovine Serum Albumin (BSA) in PBS-Tween 0.01%). Then membranes were
   incubated with 1:1000 primary antibodies, rabbit anti-Caspase 3, or
   mouse anti-β-actin (Cell Signaling Technology #9662 and #3700,
   respectively), and rabbit anti-BRCA1-associated RING domain protein 1
   (BARD1) (Novus Biological #NB100-319) and incubated overnight at 4 °C
   with shaking. Membranes were washed and incubated with 1:20,000
   secondary antibodies anti-mouse Alexa fluor 680 (Invitrogen: #A32729)
   or anti-rabbit Alexa Fluor 800 (Invitrogen #A32735). Finally, membranes
   were washed, and the fluorescence signal was detected using Odyssey
   CLx. Analysis was made using Image Studio software (version 5.2).

JC-1 mitochondrial membrane potential assay

   To assess the MMP, 500,000 PBMCs were seeded on 48-well plates and
   mitocepted with UC-MSC-derived MT previously labeled with MitoTracker
   Deep Red. The next day, cells were treated with 10 µM or 20 µM STS for
   5 h to induce apoptosis. Then, cells were collected and stained with
   anti-CD3 PECy7 antibody for 30 min at 4 °C in darkness, washed once
   with PBS, and labeled with 2 μM of JC-1 (Invitrogen #[84]M34152) for
   30 min at 37 °C and 5% CO[2] conditions. Finally, cells were collected
   and analyzed immediately using FACS Canto II and FlowJo v.10 software.

RNA extraction and quantitative RT-PCR

   For qRT-PCR, total RNA was isolated from 1 × 10^6 FACS-sorted CD3^+
   MitoT^pos or CD3^+ MitoT^neg cells, as previously reported [[85]17].
   Complementary DNA was prepared in a reverse-transcription reaction
   using 600 ng of RNA and SuperScript II Reverse Transcriptase
   (ThermoFisher Scientific #18064014). 5 ng of template mRNA and 5 µM of
   each forward and reverse human specific primers (Additional file [86]1:
   Table S1) were used. qRT-PCR was performed in duplicate on a Stratagene
   Mx3000P using Brilliant II SYBR Green Master Mix (Agilent Technologies
   #600828). Data were expressed as relative gene expression using the
   2−ΔCt method and normalized to 18S rRNA housekeeping gene.

Target cell killing assay

   To analyze the cytotoxic potential of the generated cells, an in vitro
   assay was conducted using effector cells against the NALM-6 GFP+ target
   cells. Prior to the assay, the number of CAR+ cells was verified for
   all conditions, and the percentage of CAR-T cells was normalized across
   each donor to ensure uniformity (ranging from 8 to 1% CAR+ cells per
   donor). The respective donor’s Mock (or Mock MitoT^pos) was used for
   this normalization. After 8 days of cell expansion, two different
   effector-to-target ratios were plated in U-bottom plates: 1:1 (30,000
   cells each for effector and target) and 1:2 (15,000 effector cells and
   30,000 target cells). Cytotoxicity was assessed at 0 h, 48 h, and 96 h
   after co-culture by counting the remaining target cells. After each
   incubation period, the cells were washed and stained with the
   eBioscience™ Fixable Viability Dye eFluor™ 780 (#65-0865-18), and the
   samples were analyzed by flow cytometry. To quantify NALM-6 GFP+ cells,
   dead cells were excluded and the number of GFP+ counts was plotted on a
   graph to evaluate the lytic potential of the effector cells.

ELISA assay

   A co-culture assay was conducted as described in the cytotoxicity assay
   section. The 1:1 effector-to-target ratio cultivation was performed,
   and the supernatant was collected 24 h later. The ELISA of IL-2, IFNγ,
   and TNFα cytokines, the R&D Systems DuoSet kits from catalogs DY202,
   DY285B, and DY210 were used, respectively. The protocol was followed
   according to the datasheet specifications, except for the volume
   reduction in a Corning® 96-Well Half-Area Microplate. In brief, the
   plate was incubated overnight with 25 µL of the capture antibody
   solution at the protocol-recommended concentration. Subsequently, 25 µL
   of the sample’s supernatant was added and incubated overnight. The next
   day, the detection antibody and necessary reagents for revelation were
   added, and all washing and blocking steps were performed as specified.
   The plate was read on the SpectraMax iD3 at a wavelength range of
   450 nm (with 540 nm correction), and quantification in pg/mL was
   carried out according to the standard curve and blank.

Statistical analysis

   In vitro tests were performed using different biological replicates in
   at least three independent experiments. The data were analysed using
   GraphPad Prism 10.1.1 software (GraphPad Software, San Diego, CA, USA),
   and presented as mean ± SEM. Normal distribution was assessed with the
   Shapiro‒Wilk test, and no outliers were removed using ROUT method
   (Q = 1%). Differences were assessed by two-tailed unpaired Student’s
   t-tests, paired t-test or ANOVA test, as indicated for each figure
   legends, and considered statistically significant for p values of
   < 0.05. A heatmap was generated using statistically differentially
   expressed (DE) genes showing an absolute fold change (FC) > 1.5 and a
   p-value < 0.05.

Results

Transfer of UC-MSC-derived MT modifies T lymphocyte response to stimulus and
apoptosis gene expression

   Considering the central role of MT in the control of apoptosis, we
   assessed whether the direct transfer of active and functional
   UC-MSCs-derived MT [[87]17, [88]29] regulates the T cell apoptotic
   response. We performed whole transcriptome RNA-Seq analysis on
   previously reported FACS-sorted CD3^+ T cells that acquired or not
   exogenous UC-MSC-derived MT (CD3^+ MitoT^pos and CD3^+ MitoT^neg cells,
   respectively) [[89]17]. Volcano plot analysis derived from FPKM values
   depicted 205 DE protein-coding genes underexpressed and overexpressed
   between MitoT^pos versus MitoT^neg CD3^+ cells (Fig. [90]2A), and
   pathway enrichment analysis (PEA), using the Reactome tool, revealed
   the biological functions that were denoted in these genes
   (Fig. [91]2B). Among the enriched pathways, GO enrichment analysis
   revealed 20 gene terms significative enhanced, related to apoptosis,
   cell death and/or responses to different stimulus in MitoT^pos cells
   (Fig. [92]2C), and the specific DE genes contained within these GO
   terms were represented as an unsupervised hierarchical clustering
   heatmap, showing up or down-regulated transcripts between MitoT^pos
   from MitoT^neg CD3^+ sorted samples (Fig. [93]2D).

Fig. 2.

   [94]Fig. 2
   [95]Open in a new tab

   Transfer of UC-MSC-derived MT modifies T lymphocyte response to
   stimulus and apoptosis gene expression. A Volcano plot analysis derived
   from FPKM values showing DE genes (p < 0.05, Student’s t test). Red
   dots: underexpressed genes (FC MitoT^pos/MitoT^neg < − 1.5). Blue dots:
   overexpressed genes (FC MitoT^pos/MitoT^neg > 1.5). B Pathway
   enrichment analysis (PEA) for DE genes between CD3+ MitoT^pos (n = 4)
   and CD3+ MitoT^neg (n = 4) cells. C Significantly enriched GO terms
   related to apoptosis, cell death, and/or responses to different stimuli
   of DE genes (p < 0.05, n = 4) between MitoT^pos and MitoT^neg cells. D
   Unsupervised hierarchical clustering heatmap of differentially
   expressed genes associated with GO terms from MitoT^pos (n = 4) and
   MitoT^neg (n = 4) samples, from whole transcriptome RNA sequencing
   analysis. Red transcripts represent genes directly involved in the
   apoptotic process regulation

MitoT from MSC-derived MT grants apoptosis resistance to T cells

   To further assess the functional impact of MSC-derived MT on human
   T-cell apoptotic response, isolated MT from UC-MSCs were transferred to
   PBMCs through the mitoception protocol, and 24 h after, the cells were
   treated with STS, an inhibitor of protein kinase C (PKC), for 4 h to
   induced apoptosis (Fig. [96]3A). Cell viability was evaluated by
   Annexin V/7-AAD assay and flow cytometry gating on CD3^+ T population
   (Fig. [97]3B). Our results showed that CD3^+ MitoT^pos cells were
   significative more resistant to STS-induced apoptosis than MitoT^neg
   cells at either 10 μM or 20 μM of STS (18.4% versus 30.7%, and 26.0%
   versus 48.0%, respectively; Fig. [98]3C). Further analysis of apoptotic
   cell-death by TUNEL staining, corroborated that CD3^+ MitoT^pos cells
   were more resistant to STS-induced apoptosis than MitoT^neg cells, with
   a 1.9-fold decrease in the percentage of TUNEL^+ cells at 5 µM of STS
   (Fig. [99]3D and Additional file [100]1: Figure S2). Finally, caspase-3
   activation was evaluated by western blot by the ratio cleaved caspase
   3/pro-caspase 3. We observed that MitoT^pos population showed a
   significative 1.7-fold decrease of cleaved caspase-3 compared with
   MitoT^neg T cells upon STS treatment (Fig. [101]3E, [102]F),
   demonstrating the protective role of MitoT from UC-MSCs, preventing the
   STS-induced apoptosis of CD3^+ T lymphocytes.

Fig. 3.

   [103]Fig. 3
   [104]Open in a new tab

   MitoT from MSC-derived MT grants apoptosis resistance to T cells. A
   Experimental design of artificial MitoT (mitoception) used for
   apoptosis assays. B Representative FACS plots of Annexin V-apoptosis
   assay on CD3+ MitoT^pos and MitoT^neg populations after 4 h of
   staurosporine (STS) treatment. C Percentage of apoptotic CD3+ MitoT^pos
   cells compared to MitoT^neg cells after STS treatment. The left bars
   represented no STS-treatment control (n = 3). D Representative TUNEL
   assay (left panel) and percentage of TUNEL+ cells (right panel), by
   FACS, of CD3+ MitoT^pos cells compared to MitoT^neg cells after STS
   treatment (n = 3). E Representative western blot of caspase-3 cleavage
   on MitoT^pos and MitoT^neg CD3+ T lymphocytes after 4 h of STS
   treatment. β-Actin was used as loading control. F Quantification of
   proteins showed in E, represented as caspase 3/pro-caspase-3 ratio in
   CD3+ MitoT^pos cells compared to MitoT^neg cells (n = 3). G
   Representative FACS plots of the MMP by flow cytometry using JC-1
   labeling of T lymphocytes after STS treatment. H Mitochondrial
   depolarization, as the ratio of red to green fluorescence, in CD3+
   MitoT^pos cells compared to MitoT^neg cells after STS treatment by FACS
   analysis (n = 3). For figures C, D, F and H: graphs show mean ± SEM and
   statistical analysis by unpaired t-test. All replicates are biological

   Since the external MT membrane permeabilization is key in the ongoing
   downstream apoptosis events, we evaluated the changes in the MMP of T
   lymphocytes exposed to STS (Fig. [105]3G). Interestingly,
   UC-MSC-derived MitoT prevented MT depolarization when comparing
   MitoT^pos and MitoT^neg CD3^+ T cells after STS treatment (red/green
   fluorescence ratio of 0.67 versus 0.19 at 10 μM STS (p = 0.009), and
   0.60 versus 0.10 at 20 μM STS (p = 0.018); Fig. [106]3H). These results
   point to a protective role of MitoT in preventing STS-induced apoptosis
   by avoiding the dissipation of the MMP.

MitoT increases the expression of anti-apoptotic Bcl-2/BARD1 pathways

   To understand the mechanism leading to MitoT-mediated T cell apoptosis
   resistance, we performed further analysis of the RNA-seq data of the
   CD3^+ MitoT^pos and MitoT^neg cells. Fold change of the DE genes
   associated with DNA repair and apoptotic process, regulated after
   UC-MSC-derived MitoT, revealed that the most differentially upregulated
   gene was BARD1 (Fig. [107]4A), involved in the regulation of apoptosis
   and DNA repair [[108]33, [109]34]. qRT-PCR analysis confirmed the
   upregulation of BARD1 transcript levels in FACS-sorted CD3^+ MitoT^pos,
   compared to CD3^+ MitoT^neg cells (p = 0.052; Fig. [110]4B). Protein
   expression levels of full length BARD1 were also significantly
   increased in MitoT^pos sorted CD3^+ cells (Fig. [111]4C), with a
   2.1-fold increase (p = 0.033) at 24 h after MitoT (Fig. [112]4D).

Fig. 4.

   [113]Fig. 4
   [114]Open in a new tab

   MitoT increases the expression of anti-apoptotic Bcl-2/BARD1 pathways.
   A Fold change of differentially expressed (DE) genes contained within
   the GO terms related to apoptosis, cell death, and/or responses to
   different stimuli. B qRT-PCR analysis of BARD1 mRNA expression levels
   in FACS-sorted CD3+ T cells after 24-h post-mitoception with MSC
   derived-MT (n = 4). C Representative western blots of BARD1 in
   FACS-sorted CD3+ MitoT^pos cells and MitoT^neg cells after 24 or 48 h
   post-mitoception. D Fold change quantification of proteins showed in C
   (n = 3). E Ingenuity Pathway analysis for Bcl-2 gene network for the DE
   genes denoted in A. F qRT-PCR of Bcl-2 family gene expression in
   FACS-sorted CD3+ MitoT^pos cells and MitoT^neg cells after 24 h
   post-mitoception (n = 4). For figures B and D: graphs show mean ± SEM
   and statistical analysis by unpaired t-test. For figure F: graph shows
   mean ± SEM and statistical analysis by unpaired Mann–Whitney test
   (*p < 0.05, relative to MitoT^neg control). All replicates are
   biological

   Different anti- and pro-apoptotic proteins belonging to the Bcl-2
   family regulate the drop of the MMP and cytochrome c release into the
   cytosol by inhibiting or inducing the formation of pores on the
   mitochondrial membrane. Since we observed a preventive effect on the
   MMP dissipation in MitoT^pos T lymphocytes, we further analyzed the
   Bcl-2 gene network for the DE gene expression data regulated in
   MitoT^pos CD3^+ cells. Using ingenuity pathway analysis, we identify
   BARD1 and PARK2 transcripts directly related to the Bcl-2 family
   (Fig. [115]4E). Finally, qRT-PCR analysis confirmed a significative
   7.4-fold increase of Bcl-2 mRNA transcript levels in CD3^+ MitoT^pos,
   compared to MitoT^neg cells (p < 0.05). However, no changes were found
   in other Bcl-2 family members (Fig. [116]4F).

Anti-apoptotic effect on CAR-T cells

   It has been shown that during CAR-T cell production, especially with
   the electroporation process, cell viability decreases ~ 20% [[117]10].
   Therefore, since in our experiments MitoT from UC-MSC-derived MT
   increased apoptosis resistance potential to CD3^+ T lymphocytes, we
   evaluated whether MitoT could protect T cells from apoptosis induced by
   the electroporation with PiggyBac carrying a c-Myc-tagged version of
   the 19BBz CAR construct. PBMCs from healthy donors were mitocepted with
   previously MitoView-labeled MT, electroporated with PiggyBac 19BBz, and
   cell viability of CAR-T engineered cells was evaluated by Annexin
   V/7AAD assay (Fig. [118]5A). We first observed that the transduction
   efficiency of CAR-T cells was not affected by the MitoT, as a similar
   percentage of CD3^+ MitoT^neg CAR-T[pos] and CD3^+ MitoT^pos CAR-T[pos]
   cells were found after electroporation (Fig. [119]5B and Additional
   file [120]1: Figure S3A). Interestingly, our results showed that MitoT
   significantly increased the cell viability for either CAR-T[neg]
   (average of 80.1% and 10.2% respectively, p < 0.0001) and CAR-T[pos]
   (average of 52.2% and 5.6% respectively, p = 0.0053) populations, when
   comparing MitoT^pos and MitoT^neg cells (Fig. [121]5C, [122]D).
   Consistently, we were able to confirm the anti-apoptotic effect of
   MitoT, since a significant reduction of early and late apoptosis was
   observed in MitoT^pos compared to MitoT^neg subpopulations for both
   CAR-T[pos] and CAR-T[neg] cells (Fig. [123]5C). The results showed a
   4.2-fold and 17.2-fold reduction of early and late apoptosis in
   CAR-T[neg] populations (average of 14% and 1.7% for MitoT^pos; and
   58.2% and 29.3% for MitoT^neg) (p = 0.004 and p = 0.0039,
   respectively). Similarly, a 1.7-fold decrease and 2.4% reduction of
   early and late apoptosis were observed in CAR-T[pos] populations
   (average of 27.0% and 19.2% for MitoT^pos; and 46.2% and 45.9% for
   MitoT^neg) (p = 0.0344 and p = 0.0197, respectively; Fig. [124]5D). No
   differences in necrotic subpopulations were observed (Fig. [125]5C,
   [126]D). These data showed that MitoT protects CD3^+ lymphocytes,
   especially CD3^+ CAR-T engineered cells, from electroporation-induced
   apoptosis.

Fig. 5.

   [127]Fig. 5
   [128]Open in a new tab

   Anti-apoptotic effect on cMyc-tagged CAR-T cells. A Experimental design
   of artificial MitoT used for apoptosis assay on CAR-T cells. B
   Representative plots (left panel) and percentage of CAR[pos] cells
   (right panel), by FACS, of CD3+ MitoT^pos cells compared to MitoT^neg
   cells after cell-electroporation (n = 4). C Representative FACS plots
   of Annexin V-7AAD apoptosis assay on CD3+ MitoT^pos and MitoT^neg cells
   in CAR[neg] and CAR[pos] populations after cell-electroporation. D
   Percentage of live, apoptotic, and necrotic CD3+ MitoT^pos and
   MitoT^neg cells in CAR[neg] and CAR[pos] populations after
   cell-electroporation (n = 4). E, F Mean of cytotoxicity activity of
   lymphocytes mitocepted (MitoT^pos) or not (MitoT^neg) and
   electroporated with 19BBz or Mock control, against Nalm-6-GFP B-cell
   line, after 8 days of expansion. The effector cells were incubated in a
   1:2 (E) or 1:1 (F) ratio with the target cells (E:T) for 48 or 96 h.
   Right bar graph shows the average of remaining target cells (counts)
   (n = 4). G Representative plots (left panel) and percentage of CAR[pos]
   IFNγ^+ cells (right panel), by FACS, of CD3^+ lymphocytes mitocepted
   (MitoT^pos) or not (MitoT^neg) after 8 days of expansion (n = 4). H
   Cytokines quantification, by ELISA, on supernatant from lymphocytes
   expanded and co-cultured for 24 h with target cells (1:1 ratio)
   (n = 4). Graphs show mean ± SEM and statistical analysis by unpaired
   t-test for figures B and D, paired t-test for figures E, F and G, and
   One-way ANOVA for figure H. All replicates are biological

   Additionally, after CAR-T production by electroporation, the effector
   function of expanded T cells was evaluated. We observed that 19BBz
   MitoT^pos cells showed an increasing trend in cytotoxic activity
   against Nalm-6 B cell leukemia after 96 h of co-culture, compared to
   19BBz MitoT^neg cells, although this difference was not significtive
   (Fig. [129]5E, [130]F). Curiously, MitoT^pos CAR-T[pos] cells displayed
   a significantly lower intracellular IFNγ levels than MitoT^neg
   CAR-T[pos] population (p = 0.042) (Fig. [131]5G and Additional file
   [132]1: Figure S3B, C), but no differences were observed on cytokines
   released in the supernatant (Fig. [133]5H and Additional file [134]1:
   Figure S3D). Finally, the assessment on exhaustion markers (LAG3, PD1,
   and TIM3), memory markers (CD95+, CD45RO, and CCR7), and Treg CD4+
   expression, showed no differences, neither between the population
   CD4+CAR+CD4^+ CAR-T[pos] MitoT^pos compared to CD4^+ CAR-T[pos]
   MitoT^neg, or between CD4^+ CAR-T[neg] MitoT^pos compared to CD4^+
   CAR-T[neg] MitoT^neg (data not shown).

Discussion

   There is growing interest in using MT as a therapeutic agent to treat
   inflammatory and degenerative diseases. Recent studies revealed that T
   cells can receive MT from other cell types, affecting T cell
   development and/or function. As previously demonstrated by our group, a
   range of immunocompetent cells, including T and B cells, are receptive
   for MT of different MSC-sources. Such MitoT has an effect on T cell
   metabolism, promoting T[reg] cell differentiation and improvement in
   tissue damage and survival in a GVHD model of Disease [[135]17].
   Similarly, MitoT from MSCs to CD4+ T helper type 17 (Th17) cells
   promotes metabolic changes, reducing the expression of pro-inflammatory
   cytokine IL-17 [[136]35], while the transfer of intact MT into
   rheumatoid arthritis (RA)-associated T cells restores metabolic defects
   conferring anti-inflammatory properties that reduce disease severity in
   a humanized mouse model [[137]36]. These data suggest that exogenous MT
   appear to metabolically reengineer target T cells that might display
   increased therapeutic potential approach. To this end, it is critically
   important to understand the underlying rationale and available evidence
   that support the use of MT as a cell-free product in ex vivo procedures
   or in clinical settings.

   MT plays a key role in regulating apoptosis in mammalian cells, and its
   intracellular transfer has been proposed as one of the mechanisms for
   apoptosis prevention of damaged cells. Indeed, Wang et al. exposed that
   TNT-mediated transfer of functional MT reversed cellular stress in
   early apoptotic stages in pheochromocytoma (PC) 12 cells [[138]37]. In
   a separate approach, the microinjection of intact MT into oocytes
   reduced their high susceptibility to apoptosis [[139]38].

   In the present study, we report that the transfer of isolated active
   and functional MT from UC-MSCs changes the expression of mRNA
   transcripts involved in T cell response to stimulus, cell death and
   apoptosis, including PLD1, BARD1, TNFR2 and DUSP1. We also demonstrated
   the increase resistance to STS-induced apoptosis in CD3^+ T cells that
   acquired MT from UC-MSCs (CD3^+ MitoT^pos cells). While these results
   are consistent with other group reports showing the anti-apoptotic
   effect of MSCs-derived MT, no study has shown the role of MitoT on the
   survival of apoptotic-challenged immunocompetent cells. In tenocytes
   injured by H[2]O[2], MitoT rescues distressed cells from apoptosis, and
   prevents Achilles tendinopathy in vivo [[140]39]. Additional work has
   shown that the active MitoT from adult MSCs to mtDNA-deficient cells
   increases survival and restores their MT function [[141]40]. Also, in
   odontoblasts, bone marrow MSCs (BMSC)-derived MitoT protects from
   pyroptotic cell death in response to dental pulp damage [[142]41].

   The selection of an appropriate tissue source for MSCs used in the
   isolation of MT requires a balancing of considerations, including cell
   yield, potency and clinical testing. We have selected the umbilical
   cord source due to its extensive use in clinical trials, supported by
   our group’s experience with UC-MSCs across various completed clinical
   applications [[143]28, [144]42–[145]44].

   Preservation of the permeability of the MT outer membrane is crucial
   for cell survival. It is known that one of the main pathways of
   apoptosis involves Caspase-3 which, when activated, catalyzes the
   cleavage of numerous key cellular proteins and changes the MT membrane
   permeability and swelling [[146]45]. Here, when caspase-3 activation
   was measured following protein cleavage, CD3^+ MitoT^pos cells
   displayed decreased levels of cleaved caspase-3 compared with the
   MitoT^neg population. Additionally, since the decline in MMP and
   further release of cytochrome c from MT are key events during
   apoptosis, we assayed the effect of MitoT on MT membrane depolarization
   in apoptotic-challenged T lymphocytes. We observed that UC-MSC-derived
   MitoT prevented the reduction of MMP induced by STS in CD3^+ MitoT^pos
   cells, reinforcing the protective role of MitoT in STS-induced
   apoptosis. Furthermore, Anti-apoptotic Bcl-2 family members have been
   shown to inhibit the release of cytochrome c and loss of the MMP
   associated with apoptosis [[147]46]. Within our experimental setting,
   an increased expression of the anti-apoptotic gene Bcl-2 was detected
   in FACS-sorted MitoT^pos CD3^+ cells, suggesting that the prevention of
   the drop in the MMP might be mediated by the increase in the Bcl-2
   expression.

   These results present a new approach to increase the survival advantage
   of T cells and improve their resistance to environmental stress
   following their transplantation. For example, Tschumi et al. showed
   that CAR-T cells progressively upregulate Fas, FasL, DR5, and TRAIL,
   which result in their programmed cell death independently of their
   antigen-mediated TCR or CAR activation [[148]47]. Improving the
   fitness, survival, and persistence of T cells used in immunotherapy has
   important translational value because they constitute key factors when
   assessing an effective therapy outcome. As an example, adoptive cell
   therapy of T[reg] cells for the treatment of autoimmune diseases also
   encounters limitations given the ex vivo expansion of T[reg] cells is
   patient-dependent. In addition, they require IL-2 infusion to support
   cell survival and also patients under this treatment are often in need
   of further immunosuppressant drugs that limit both survival and
   proliferation of T[reg] cells [[149]48].

   It is noteworthy that the observed protective effect is closely
   associated with alterations in cellular metabolism, enhanced ATP
   production, improved cell proliferation, and reduced ROS levels,
   thereby mitigating mechanical and oxidative stress-induced apoptosis.
   This may also accelerate metabolic reprogramming, promoting an
   activated phenotype with increased mitochondrial biogenesis and OXPHOS
   capacity. This could potentially enhance therapeutic outcomes of CAR-T
   cell therapy, by improving metabolic fitness and resistance to
   apoptosis.

   Interest in CAR-T therapy has increased exponentially in recent years
   as clinical applications seem to expand from hematologic cancers
   [[150]49] to autoimmune refractory disease [[151]50]. Specifically, the
   ability of MitoT to prevent apoptosis in T lymphocytes and CAR-T cells
   suggests a potential therapeutic application in enhancing the efficacy
   and longevity of CAR-T cell therapies. The latest evidence enphasis the
   significance of T cell metabolism in cancer therapy, underscoring the
   necessity for metabolic modulation in CAR T cell therapies for an
   improved clinical outcome [[152]51].

   Most CAR-T-cell approved or in clinical trials require a viral
   transfection to induce a genetic modification, which is associated with
   high production costs, increased production time, and cumbersome
   quality controls that are reflected in the final cost of CAR-T therapy.
   Therefore, technologies that reduce manufacturing time and costs, while
   maintaining T-cell function are a requisite to expand the usage of CAR
   T-cell therapy. New approaches with virus-free and in vivo gene
   delivery systems offer promising avenues to this effect [[153]52].
   Several groups described the use of Sleeping Beauty (SB) transposons
   for CAR T-cell generation [[154]53, [155]54]. As shown by Chicaybam et
   al., efficient genetic modification of T lymphocytes with high and
   long-term CAR expression was induced with the SB transposon system,
   displaying antitumor activity in vitro and in vivo [[156]10]. Moreover,
   the authors also describe a protocol to generate 19BBz CAR-T cells in
   8 days, with clinical-grade applications, developed to meet the
   increasing demand for this therapy [[157]11]. Clinical trials applying
   electroporation to generate CAR-T cells using non-viral vectors such as
   Sleeping Beauty or PiggyBac are underway [[158]55–[159]58]. In
   addition, electroporation results support good clinical results and
   therapeutic potential for CAR-T cells using this platform [[160]59,
   [161]60]. Nevertheless, electroporation-induced apoptosis of T cells is
   a major limit to the efficacy and broad application on the advantages
   of producing CAR-T cells with non-viral transfection. In this work, we
   evidenced that MitoT confers a reduction of early and late apoptosis,
   as observed for both CAR-T[pos] and CAR-T[neg] subpopulations. Although
   most of the clinical applications with transposons were performed with
   SB-based transposons, whenever a large cargo of transgene is considered
   (∼ 200 kb versus ∼ 10 kb), PiggyBac might be the transposon of choice.
   In this regard, here we validate the anti-apoptotic effect of MitoT in
   CAR-T cells electroporated with the non-viral transposon-based vector
   PiggyBac.

   CAR-T therapy has potentially severe and hard-to-manage side effects,
   such as cytokine release syndrome (CRS), neurotoxicity, and macrophage
   activation syndrome. IFNγ increases CAR-T cytotoxicity and the
   antitumoral response. However, higher levels IFNγ are related with CRS
   and neurotoxicity [[162]61]. A recent study evaluated a strategy of
   neutralizing IFNγ with the monoclonal antibody emapalumab that
   mitigates severe toxicity related to CAR-T cell therapy [[163]62]. Our
   results showed that MitoT reduced the expression of IFNγ, displaying an
   increasing trend in cytotoxic activity against Nalm-6 cells.

   Since the long-term persistence, higher cytotoxicity, with lower side
   effects of CAR-T cells is one of their advantageous traits, we plan to
   address the antitumor capacity, persistence, and CRS of MitoT CAR-T in
   future studies with humanized animal models to confirm that these
   strategies could lead to improved and more affordable CAR-T cell
   therapy.

   The findings presented in this research broaden the physiological
   perspective of organelle-based therapies in immune conditions where the
   regulation of apoptosis and cell survival is diminished.

Conclusions

   We have described a model for the assessment of the protective effect
   of MitoT on a variety of cell types while showing an acquired survival
   advantage in specific T and CAR-T cell populations. Our data suggest
   UC-MSC-derived MitoT regulates CD3^+ T cell apoptosis in a
   BCL2-BARD1-dependent manner, leading to a marked functional impact on T
   lymphocyte survival. The transfer of external MT might offer an avenue
   to potentiate the therapeutic effects of adoptive transfer of
   unmodified T or virus-free based protocols for generating CAR-T cells.

Supplementary Information

   [164]Supplementary Material 1.^ (727.8KB, docx)

Acknowledgements