Human leukemic cell lines CII (#ACC 773), MEC-1 (ACC 497), and the megakaryocytic Set2 cell line (#ACC 608) were purchased from the Leibniz Institute DSMZ-German Collection of Microorganisms and Cell Culture (DSMZ Braunschweig, Germany). The cells were grown and maintained in RPMI 1640 media (ThermoFisher Scientific, Burlington, ON, Canada) supplemented with 10% fetal bovine serum/FBS (Wisent Bioproducts, St‐Bruno, QC, Canada), FBS) at 37°C under 5% CO₂ in a humified incubator. Set2 cells were grown in RPMI 1640 medium supplemented with 20% fetal bovine serum. Culturing of cells was performed at a medium density of 2-5 x 10⁵ cells/mL for CII and 0.5-2.0 x 10⁶ cells/mL for MEC-1 and split at an appropriate ratio of 1:2 every 40 hrs. Cell treatments for phenotypic analyses included anticancer drug compounds Cytarabine, Venetoclax, and Plumbagin (Sigma-Aldrich) at the indicated concentrations for the indicated time points. To evaluate PMP internalization levels into cells, co-cultures were performed with cells labeled with anti-human CD19-FITC antibody (BioLegend, San Diego, CA, USA) and MitoTracker^® Deep Red (MTDR, ThermoFisher Scientific)-labeled PMPs followed by cell cytometry using the 644 nm EX for MTDR and 488 nm EX for CD19 detection.

This study was conducted in accordance with the guidelines of the Declaration of Helsinki of 1975 (revised 2013, https://www.wma.net/what-we-do/medical-ethics/declaration-of-helsinki/). Institutional Review Board Statement and approval for studies involving human samples have been granted from both the Vitalité Health Network (project: CER7-3-17 Ver. 5) and the Université de Moncton (NB, Canada) (project: 1516-002) research ethics boards. Peripheral blood was obtained from consenting healthy volunteers (three men and seven women, aged 22-37). Healthy donors selected did not have any known autoimmune diseases and did not previously receive any preoperative chemotherapy and/or radiotherapy.

Patient samples consisted of 200 mL of whole peripheral blood, collected from cubital vein using a 21-gauge needle and Vacutainer tubes (ThermoFisher Scientific) containing the anticoagulant acid citrate dextrose (ACD; 0.085 mM sodium citrate, 0.0702 mM citric acid, 0.111 mM dextrose, pH 4.5). The platelets-rich plasma (PRP) layer was separated from the erythrocytes and other cell types by centrifugation at room temperature for 10 min at 265 x g. The PRP was transferred to a new tube where ACD (1:5 of the initial PRP volume) and ethylenediaminetetraacetic acid (EDTA, 10 mM, pH 8.0) were added, and the mixture was subsequently centrifuged at 400 x g for 2 minutes at room temperature. The supernatant was collected into a new tube and centrifuged at 1,300 x g for 10 minutes at room temperature. The harvested pellets were gently suspended in 200 μL Tyrode buffer (134 mM NaCl, 2.9 mM KCl, 0.34 mM Na₂HPO₄, 12 mM NaHCO₃, 20 mM HEPES, 1 mM MgCl₂, 5 mM glucose, and 0.5 mg/mL bovine serum albumin (BSA), pH 6.5) and then transferred in 25% volume of initial PRP of Tyrode solution (pH 7.4). Platelet count was performed with a hemocytometer slide under a Nikon TMS inverted phase-contrast microscope equipped with a Coolpix 5000 high-resolution digital camera (Nikon Corp, Melville, NY, USA). PMP production was initiated by incubating platelets (3 x 10⁸ cells/mL) with thrombin (0.1 U/mL, Sigma-Aldrich, Oakville, ON, Canada) and 5 mM CaCl₂ overnight at room temperature. The next day, EDTA (10 mM) was added to stop the reaction and remaining platelets were isolated by centrifugation at 500 x g for 10 min. The supernatant was collected and centrifuged at 17,800 x g for 90 min at room temperature. Pelleted PMPs were then resuspended in phosphate-buffered solution (PBS, Sigma-Aldrich).

PMP quantification were performed by platelet fluorescence labeling with either 100 nM MTDR (mitochondria staining) with 2 μM PKH67 (Sigma-Aldrich) (cell membrane staining), or anti-CD41-FITC (BioLegend) (membrane staining), followed by an incubation for 15 min at 37°C. The labeling process was stopped by adding 10 mM EDTA and subsequent centrifugation at 1,300 x g for 10 min. The pellet, containing labelled PMPs, was resuspended in Tyrode buffer and characterized by flow cytometry (Attune NxT Flow Cytometer, ThermoFisher Scientific) using the 644 nm excitation (EX) and 665 nm emission (EM) for MTDR and 490/502 nm for PKH67. All solutions were filtered with a 0.2 µm filter before use. All experiments were performed with freshly isolated PMPs and were never frozen to ensure the functional integrity of mitochondria, unless otherwise indicated.

Mitochondrial isolation was performed using a PIERCE Mitochondria Isolation Kit (ThermoFisher Scientific) according to the manufacturer’s instructions. A total of 2 × 10⁷ cells were pelleted by centrifuging at 850 × g for 2 min and subsequently treated with 800 μL of Mitochondria Isolation Reagent A, vortexed and kept for 2 min on ice. The samples were then mixed with 10 μL of Mitochondria Isolation Reagent B and chilled for 5 min. At the end of incubation, cells received 800 μL of Mitochondria Isolation Reagent C and centrifuged at 700 × g for 10 min at 4°C. The supernatant was then collected and centrifuged at 12,000 × g for 15 min at 4°C. The obtained pellet was mixed with 500 μL of Mitochondria Isolation Reagent C and centrifuged again at 12,000 × g for 5 min. The purified mitochondria were kept on ice before downstream handling.

Upon isolation, mitochondria were labeled with MTDR and quantified using flow cytometry. Purified mitochondria were then mixed with PBS and co-cultured with cells labeled with DAPI and MitoTracker™ Green FM (MTG, ThermoFisher Scientific) at 10 mitochondria per cell (10:1), and 100 mitochondria per cell (100:1) ratios as previously instructed (48, 49). Following a 24- and 48-hour incubation, cell media was replaced with fresh RPMI-1640 supplemented with 10% FBS. Cells were monitored using an Olympus FV1000 confocal fluorescence microscope at 600x magnification (UPLAN 60x oil, 1.35NA, Olympus) with appropriate excitation laser and filters. Confocal microscopy fluorescence was then confirmed by flow cytometry where gating events were based on CD19 expression and MTDR-labeled mitochondria internalization.

Cell viability was measured by the Cell Titer Blue assay (Promega, Madison, WI, USA) based on the manufacturer’s guidelines. Briefly, cells were seeded in a 96-wells plate at a density of 5 x 10³ cells/well and treated with PMPs (10 PMP/cell) for the indicated time points. Thereafter, 20 μL of Cell Titer Blue substrate was added to each well and incubation under standard cell culture conditions for 1 hour. The plate was read with a Synergy H1 Hybrid Multi-Mode Microplate Reader (Agilent, Santa Clara, CA, USA) at Ex/Em: 560/590 nm via Gen5 data analysis software (Version 2.01).

Cell-cycle progression was evaluated with a propidium iodide (PI) solution (MilliporeSigma, Burlington, Massachusetts, USA). Briefly, 5 x 10⁶ cells were collected by centrifugation at 300 x g for 5 min, washed in 1 mL PBS buffer containing 2% FBS, and resuspended in 5 mL chilled 70% ethanol dropwise and kept at 4°C. After a 24-hour incubation, fixed cells were centrifuged at 850 x g for 10 min, PBS-washed, and resuspended in a staining mixture of 1 mL PI staining solution + 50 µL RNase A (New England Biolabs, Whitby, Ontario, Canada) at 4°C overnight. Samples were then analyzed by flow cytometry at Ex/Em: 536/617 nm using the FlowJo software (Version 10.7, Tree Star, Inc., Ashland, OR., USA).

Apoptosis was measured using the FITC Annexin-V Apoptosis Detection Kit with PI (BioLegend). In accordance with the manufacturer’s protocol, 100 μL of PBS-washed cell samples were diluted in Annexin-V Binding Buffer at a final concentration of 1 × 10⁷ cells/mL and subsequently treated with 5 μL of FITC Annexin-V + 10 µL of PI solution for 15 min at room temperature in the dark. After incubation, 400 µL of Annexin-V Binding Buffer was added to each sample and then analyzed by flow cytometry.

Cellular oxygen consumption was measured using high-resolution respirometry (Oxygraph-2k, Oroboros Instruments, Innsbruck, Austria) as previously described (50). Briefly, oxygen calibration was performed at 37°C with a continuous stirring speed of 750 rpm prior to measurements of mitochondrial OCR using 2.5 x 10⁶ cells mixed with 2 mL of complete media. Evaluation of physiological ROUTINE respiration was determined as the basal state of cellular oxygen consumption (i.e., with endogenous substrates). The non-phosphorylating resting state (LEAK respiration) was obtained after injection of oligomycin (Omy; 2.5 µM) to inhibit ATP synthase. The uncoupled respiratory capacity (electron transport system capacity; ETS) was determined as the maximal capacity of the mitochondrial ETS and was measured by adding sequential injections of carbonyl cyanide m-chlorophenyl hydrazone (CCCP; 0.5 µM steps) until no further increase in OCR was detected. Residual oxygen consumption (ROX) was subsequently obtained with injections of rotenone (ROT; 0.5 µM) to inhibit complex I and antimycin A (AmA; 2.5 µM) to inhibit complex III (50, 51). All inhibitors were purchased from Sigma-Aldrich. OCRs were corrected with ROX values and analyzed using DatLab software 7.2 (Oroboros Instruments).

Cellular ATP levels were determined by an ATP determination kit (ThermoFisher Scientific) according to manufacturer’s instructions. Briefly, 1 x 10⁴ cells were seeded in 96-well plates with fresh media supplemented with 10% FBS for 1 hour. For glycolytic ATP, cells were treated for 15 min at 37°C with ROT (0.5 μM) and AmA (2.5 μM) to inhibit mitochondrial complexes I and III, respectively. Content of the wells was then collected by centrifugation at 500 x g for 5 min at 4°C, PBS-washed, and centrifuged again. Cells were lysed in 1% Triton for 10 min at room temperature and centrifuged for 5 min at 500 x g. Then, 10 μL of each cell lysate were transferred to another 96-well plate and diluted with 90 μL of the kit’s Luciferine-Luciferase reaction reagent loaded withing the injector of a Synergy H1 Hybrid Multi-Mode Microplate Reader (Agilent). Luminescence was obtained with 1 sec integration times and analyzed with the Gen5 data analysis software (Agilent).

The CellROX^® Green Flow Cytometry Assay Kit (ThermoFisher Scientific) and the MitoSOX™ Red mitochondrial superoxide indicator (ThermoFisher Scientific) were used for intracellular ROS and mitochondrial superoxide determination, respectively. To this point, 5 x 10⁵ cells/well were added to 24-well plates and treated with PMPs as described previously. N-acetylcysteine (NAC) and Tert-butyl hydroperoxide (TBHP) were used as negative and positive controls, respectively. Following co-culturing conditions, cells were PBS-washed three times and kept in 1 mL of media containing one of the following ROS inducers: i) 6 μM AmA + 5 μM ROT; ii) phorbol 12-myristate 13-acetate (PMA, 80 ng) for 15 min; or iii) mixture of AmA + ROT + PMA for 30 min. Then, each well received 1 μM/mL CellROX green reagent for 45 min along with 5 μM SYTOX Red Dead Cell Stain (1μl/mL) for 15 min. The samples were then analyzed by flow cytometry at Ex/Em: 508/525 nm (CellROX green) and 640/658 nm (SYTOX Red Dead Cell Stain). As for the evaluation of mitochondrial superoxide, 1 x 10⁵ cells/mL from each experimental group were mixed with 5 μM MitoSOX™ reagent working solution and kept from the light for 10 min at 37°C. The samples were then analyzed by flow cytometry at Ex/Em: 510/580 nm.

Total RNA was extracted from the cell samples using TRIzol^® reagent (ThermoFisher Scientific) and treated with DNase-I for 1 hour (Sigma-Aldrich). The integrity of RNA samples was estimated using a Nanodrop ND-1000 spectrophotometer (Nanodrop Technologies Wilmington, DE, USA). First-strand cDNA was synthesized with the SuperScript III Reverse Transcriptase (ThermoFisher Scientific) based on the manufacturer’s recommendations.

Real-time PCR analysis of target genes was carried out on a CFX Connect™ Real-Time System (Bio-Rad Laboratories, Hercules, CA, USA) with the PerfeCta SYBR Green Supermix kit (Quanta Biosciences, Beverly, CA, USA). Reaction mixtures were incubated at 95°C for 5 min, followed by 39 cycles consisting of 94°C for 15 sec, 52°C for 30 sec, and 72°C for 30 sec. Amplifications were analyzed with the Bio-Rad CFX Maestro software (Bio-Rad Laboratories) and normalized to human 18S rRNA using the 2^-ΔΔCT formula (52). Primers are shown in Table 1 .

Mitochondrial DNA (mtDNA) copy number per diploid nuclear genome was determined by relative quantitative PCR (qPCR) as previously described (53). Total DNA from 5 x 10⁶ cells was extracted from cells using the gMAX DNA Mini Kit (IBI Scientific, Peosta, IA, USA) following manufacturer’s instruction. DNA was evaluated for the expression levels of mt-tRNA (Leu) and normalized to 18S rRNA using the 2^-ΔΔCT formula, where ΔCt = (Ct _nucDNA – Ct _mtDNA) ₍ 52). Data was obtained using qPCR on a CFX Connect Real-Time PCR instrument (Bio-Rad) using the PerfeCTa^® SYBR^® Green SuperMix (Quantabio, Beverly, MA, USA) following the protocol: 95°C for 5 min along with 94°C for 15 sec, 52°C for 30 sec, and 72°C for 30 sec (x 39 cycles). Primer sequences used for mt-tRNA Leu analysis are presented in Table 1 .

To investigate the mobility and invasion of CLL cells, a total of 5 × 10⁵ cells/group was FBS-starved for 24 hours before the assay, mixed with 100 μL of media containing 2% FBS on the next day, and added to the insert chambers of a 24-well ThinCerts cell culture plate (8.0 μm pore size, ThermoFisher Scientific). For the cell invasion assay, a layer of Matrigel^® (0.5 mg/mL, Corning, Manassas, VA, USA) was added into the top transwell insert compartment to mimic the extracellular matrix. The lower plate wells were then filled with 750 μL of media containing 10% FBS. Following a 24-hour incubation at 37°C, non-migrating/non-invasive cells were removed and migrated/invaded CLLs were collected, stained with CD41 antibody for 15 min at 37°C, PBS washed and read with the flow cytometer.

Cytoscape version 3.8.1 (54) was used for network illustration. The interaction between genes was predicted by the plugin BisoGenet (55). It works by integrating the gene expression data with protein-protein interaction (PPI) data to create a comprehensive view of the interactions between genes and proteins in a given biological system using several different layout algorithms including force-directed, hierarchical, and circular layouts. CentiScaPe (Version 2.2) was applied for network topology. The list of genes involved in cancer metabolism was achieved from the DisGeNET tool (Version 7.0), a Bioinformatics tool for integrating and analyzing disorder-related genes in humans (56). DisGeNET uses databases like Comparative Toxicogenomics Database (CTD) (57), Genetic Association Database (GAD) (58), Mouse Genome Database (MGD) (59), and Online Mendelian Inheritance in Man (OMIM) (60) to identify human disease-related genes; and subsequently sorted them based on the number of sources, type of organism, and supported publications in the literature.

Expression analysis of target genes in human CLL samples was carried out using the public data sets GSE58211 (N = 287) and GSE40571 (N = 155) available on NCBI Gene Expression Omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo/, accessed December 2021). Robust multi-array average (RMA) was used to remove the local biases and normalize the intensity expression of raw microarray samples to achieve the gene expression values. Hclust R-function performed hierarchical cluster analysis in R statistical environment to sort the normalized data. Identification of differentially expressed genes (DEGs) in sample files was carried out by linear models for microarray data (LIMMA) package. Annotation of genes was performed through the EnrichR server (http://amp.pharm.mssm.edu/EnrichR, accessed January 2022). The values were reported as significant if p < 0.05.

Statistical analyses were performed with the SPSS software (Version 26, IBM Corporation). The normalization and homogeneity of variances of obtained data were done with the Shapiro-Wilk and Levene’s tests, respectively. Pairwise comparisons were made using an unpaired two-tailed Student t test. One/two-way analysis of variance (ANOVA) and Tukey’s multiple comparison tests were applied to find the statistically significant differences between multiple experimental groups as appropriate. The Pearson chi-square test was used to evaluate the distribution between categorical variables. The data are presented as the mean ± SEM (Standard Error of Mean) and deemed significant if p < 0.05.

We and others have previously established PMP interaction and internalization into various types of recipient cancer cells (25, 61). To evaluate the potential PMP-mediated phenotypic regulation of leukemic cells, we first set out to validate the interaction of PMPs with CLL cell models using fluorescence-based approaches. Human leukemic cell lines (CII and MEC-1) were co-incubated with PMPs labeled with MitoTracker Deep Red (MTDR) at a ratio of 1:10 and 1:100 (cell:PMP) for 24 hours and examined for recipient cell acquired fluorescence transferred from labeled PMPs. Flow cytometry analysis demonstrated that both CLL cell lines acquired MTDR fluorescence indicating that CII and MEC-1 models acquire PMP-derived mitochondria ( Figure 1A ). Cell fluorescence from MTDR-labeled PMPs increased with cell/PMP ratios where nearly 100% of cells were positive for PMP-derived mitochondria at a 1:100 (cell/PMP) ratio after only 24 hours.

To further distinguish PMP internalization from outer cell surface interactions with target cells, we conducted fluorescence microcopy to depict intracellular compartmentalization of PMP-derived fluorescence. To do so, CII and MEC-1 cells were first stained with MitoTracker™ Green FM (MTG) and then co-incubated with MTDR-labeled PMPs at a ratio of 1:10 and 1:100 (cell: PMP) for 24 hours. Fluorescence microscopy confirmed that exogenous MTDR-labeled mitochondria provided by PMPs was transferred to CII and MEC-1 intracellular compartments. Interestingly, exogenous MTDR-labeled mitochondria also co-localized with the sub-compartmentalized region of native MTG-labeled mitochondria ( Figure 1B ).

To further validate horizontal transfer of platelet-derived mitochondria via PMPs, we performed qPCR analysis on mt-tRNA (Leu) as a mitochondrial DNA (mtDNA) or copy number indicator in CLL models co-incubated with PMPs. We observed that mtDNA was increased 1.39- and 1.91-fold in CII cells treated with 1:10 and 1:100 (cell/PMP) ratios for 48 hours, respectively ( Figure 1C ). Similar results were obtained for MEC-1 cells, which exhibited a 1.3- and 1.8-fold increase of mtDNA following treatments of PMPs with 1:10 and 1:100 ratios (48 h), respectively in comparison to control cells. Altogether, these results demonstrate that PMP cargo, notably mitochondria, are efficiently transferred into CLL cells, which result in the accumulation of mitochondrial content in recipient cells in a relatively short period of time (within 24 hours).

Given the prominent role of mitochondria in cancer cell viability and growth, we next examined the phenotypic impact of PMP internalization on recipient CLL cancer processes. Cell viability was therefore assessed in CLL co-cultures with varying ratios of PMPs over time (24 and 48 hours) using a colorimetric cell-based assay. Accordingly, PMP co-culture with CII or MEC-1 cells induced significant cell growth, which was more pronounced after 48 hours. For example, CII and MEC-1 cells treated with PMPs (1:100) increased viability 1.9- and 1.7-fold, respectively, when compared to control cells ( Figure 2A ). Interestingly, raising the ratio of PMPs in CLL co-cultures (from 1:10 to 1:100) did not significantly enhance cell viability beyond the impact observed at a 1:10 ratio at 48 hours. These observations not only suggest that PMPs play a role in leukemia proliferation, but also show that a relatively small amount of PMP interactions is sufficient to induce significant CLL cell growth.

It is well established that PMP cargo encompasses a variety of bioactive content (beyond mitochondria) capable of provoking phenotypic changes in recipient target cells (22, 23). To discern the specific role of PMP mitochondria (from other bioactive cargo) in PMP-induced CLL cell growth (above), we treated CLL models with purified exogenous mitochondria organelles ( Figure 2A ). Technically, we made use of a platelet precursor cell line (i.e., Set2 megakaryocytes) to isolate and purify mitochondria organelles intended for CLL treatments. Here, CII and MEC-1 cells were co-cultured with different ratios (1:10 and 1:100) of purified Set2-derived mitochondria over time (24 and 48 hours) and confirmed previous observations where CLL cell viability was induced with increasing amounts of purified mitochondria. Specifically, treatments of CII and MEC-1 cells with Set2-derived mitochondria (1:100 ratio) increased viability (1.6- and 1.6-fold, respectively) after 48 hours ( Figure 2B ). Control experiments to validate the intracellular compartmentalization of Set2-derived mitochondria and levels of mtDNA in CLL recipient cells were also performed following mitochondria treatment using cell cytometry, fluorescence microscopy, and qPCR ( Supplementary Figures 1A-C , respectively). Morphological analysis of CLL co-cultures (1:100 ratio) at various time points was also conducted by light microscopy ( Figure 2C ). Images of co-cultured cells show typical cell agglomeration, which is characteristic of cell growth.

Meanwhile, to better relate the direct role of mitochondria enhancement on cells proliferation, we also examined the viability of CLL cells in the presence of PMPs containing the damaged/disrupted mitochondria ( Supplementary Figure 2 ). We used two previously proven approaches where the first one consisted of freeze-thawing the PMPs prior to the co-incubation with CLLs (62). Since mitochondria are sensitive to temperature changes, cycles of freezing and thawing causes damage to their function thus impairing the motility, plasma membrane integrity and mitochondria function of boar spermatozoa through generating excessive ROS (62). The second strategy was treating the PMPs with ROT/AmA compounds, as the mitochondrial respiratory chain complex I and III inhibitors, before co-incubation with the CLL cell groups (63, 64). Upon co-incubation of leukemic cells with frozen-thawed PMPs, we observed no significant impact on cell viability in comparison to non-treated control cells ( Supplementary Figure 2A ). Similar observations were noted PMPs treated with ROT/AmA prior to co-incubation with CLL cells ( Supplementary Figure 2B ). These results showed that the proliferative impact of PMPs on cancer cells markedly depends on the health status of PMPs mitochondria. Altogether, these findings support an important contribution for PMPs in CLL proliferation. Our results also suggest that the mitochondrial component of PMP cargo is primarily responsible for accelerated growth rates of CLL cells following PMP internalization.

Mitochondrial transfer (or “mito-transfer”) is a well described phenomenon amongst cell types and has been shown to modulate recipient cell metabolic function (65–67). Since metabolic plasticity represents a hallmark of cancer progression, we investigated the impact of PMP internalization, or Set2-purified mitochondria, on several mitochondrial functions of our human leukemic cell models. First, co-cultures consisting of CLL lines (CII and MEC-1) with either PMPs or Set2-isolated mitochondria, were examined for cellular respiration through OCR using a high-resolution respirometer. We found that CLL co-cultures with PMPs (1:10) significantly increased ROUTINE (basal) respiration in both CII and MEC-1 cells (2.9- and 4.4-fold, respectively) after 48 hours when compared to controls samples ( Figure 3A ). These levels were even more pronounced (~2.7-fold higher for both cell lines) in 1:100 culture ratios ( Figure 3A ). Interestingly, co-cultures with Set2-isolated mitochondria yielded identical increases in cell ROUTINE respiration in comparison to PMP-treated cells ( Figure 3A ). These results suggest that PMP-induced basal respiration state in CLL cells is likely dependent on the transfer and function of PMP mitochondria.

We also observed that levels of non-phosphorylating resting LEAK state, after blocking the ATP synthesis with oligomycin, were also increased in PMP-treated CLL cultures, which were further accentuated with 1:100 ratios for both CII and MEC-1 cells (15.1- and 14.2-fold, respectively; Figure 3A ). These increments were reciprocated in cultures treated with Set2-purified mitochondria under the same conditions (15.9- and 17.2-fold, respectively; Figure 3A ). Finally, uncoupled respiratory capacity determined through the maximal capacity of the mitochondrial electron transport system (ETS), was also increased in PMP co-cultures, which was further accentuated in 1:100 culture ratios of CII and MEC-1 cells (7- and 13.4-fold, respectively) over control samples ( Figure 3A ). Likewise, Set2-isolated mitochondria co-cultures mimicked these increases in CII and MEC-1 cells (7.5- and 13.8-fold, respectively) in comparison to their related-control groups ( Figure 3A ).

To further elucidate the impact of PMPs on CLL mitochondrial-dependent functions, we examined ATP levels using bioluminescence assays. Total ATP levels in CII cells were increased 1.7- and 3.3-fold when cultured with PMPs (1:10 and 1:100, respectively) ( Figure 3B ). Similar ATP level increases in MEC-1 cells were also observed (2.2- and 4.1-fold) following 1:10 and 1:100 culturing ratios with PMPs, respectively ( Figure 3B ). Further investigation into the mechanistic source responsible for induced ATP levels, revealed that the OXPHOS pathway was mainly responsible ( Figure 3B ). In control experiments using Set2-derived mitochondria treatments, we observed comparable induced ATP levels suggesting that PMP-induced ATP levels in CLL cells are likely dependent on the transfer and function of PMP mitochondria.

Next, we examined intracellular ROS and mitochondrial superoxide levels in PMP-recipient leukemic cells treated with either rotenone (ROT, inhibitor of complex I), antimycin A (AmA, inhibitor of complex III), or phorbol 12-myristate 13-acetate (PMA, ROS inducer). The data depicted a significant increase in total ROS in CII cells co-cultured with PMPs (1:10 and 1:100) of 2.0- and 4.6-fold, respectively in comparison to total intracellular ROS levels from control cells alone ( Figure 3C ). To further discern the contribution of mitochondria-derived ROS production, mitochondrial superoxide was analyzed and show 1.6- and 4.8-fold increases in CII cells co-cultured with PMPs (1:10 and 1:100, respectively) when compared to their respective controls ( Figure 3D ). Similarity, MEC-1 treated with PMPs (1:10 and 1:100) induced 1.7- and 5.3-fold increments respectively, of total intracellular ROS ( Figure 3C ), where mitochondrial superoxide was increased 1.6- and 5.7-fold respectively ( Figure 3D ).

Given that Complex I is responsible for the transfer of electrons from NADH to ubiquinone (coenzyme Q) (68, 69) and that Complex III is responsible for the transfer of electrons from ubiquinone to cytochrome C (68), the combinatory use of inhibitors ROT and AmA is generally used to increase ROS levels (69). We therefore treated our cells with a ROT+AmA combination, which raised intracellular and mitochondrial superoxide levels by 3.8- and 6.4-fold in CII cells (1:100), respectively ( Figures 3C, D ). In comparison, MEC-1 intracellular ROS and mitochondrial superoxide were 3.2- and 17.2-fold higher following PMP treatments ( Figure 3C, D ). Upon ROS stimulation using PMA treatment (70–72), CII cells co-cultured with PMPs (1:100) enhanced intracellular ROS and mitochondria superoxide levels by 7.2- and 7.5-fold, respectively ( Figures 3C, D ). Likewise, PMP treated MEC-1 cells induced levels 5.1-fold (intracellular ROS) and 10.7-fold (mitochondrial superoxide), in comparison to non-treated leukemic cells ( Figures 3C, D ).

The maximum potential of ROS generation between study models was tested with mixtures of ROT+AmA+PMA. Intracellular ROS levels in CII cells treated with PMPs (1:100) increased 4.9-fold, whereas mitochondrial superoxide was raised 8.8-fold in comparison to control samples ( Figures 3C, D ). On the other hand, MEC-1 cells under these conditions raised intracellular ROS levels 4.0-fold, whereas mitochondrial superoxide increased 21.0-fold when compared to non-treated cells ( Figures 3C, D ).

Throughout these experiments, we also made used of Set2-derived mitochondria controls in parallel. We found that purified mitochondria treatments induced comparable levels of both intracellular ROS and mitochondrial superoxide as observed in PMP treatments ( Figures 3C, D ). These observations therefore suggest that total ROS production in CLL cells is likely dependent on the transfer of mitochondrial content and functions provided by PMPs. Overall, our data indicates a statistically significant increase in ROS levels, which may be attributable to either increased ROS production, or decreased antioxidant capacity of CLL cells following PMPs interaction. These results strongly suggest that PMPs enable CLL cells to gain mitochondrial-dependent functions supporting metabolic reprogramming and plasticity, which are essential for cancer malignancy.

Given the benefits from the acquisition of metabolic content and function in cancer cells, we made use of bioinformatics analyses to identify the mechanisms involved in metabolic reprogramming during disease progression and the PMP treatment. Using the BisoGenet tool, a complex network of metabolic-related genes actively involved in CLL development was developed. Network genes were then filtered based on their betweenness centrality and streamed into a smaller network consisting of 160 high-degree nodes (Hubs) where 525 edges were created as potentially key players in cancer metabolism ( Supplementary Figure 3 ). The network was then used to identify the most potential pathways encompassing these genes and highlighted: ChREBP activates metabolic gene expression (R-HSA-163765) and, the Fatty acyl-CoA biosynthesis (R-HSA-75105), as the most impacted signaling cascades in CLLs ( Figure 4A ). The expression levels of genes associated to these identified pathways were then applied to a cohort of CLL patients (n=442) based on their corresponding Binet staging (i.e., stage A (n)=184, stage B (n)=179, and stage C (n)=79) to define the most significant modulated genes associated with CLL disease progression. The average expression values of the profiled gene lists per patient are provided in Supplementary File S4 . Through our analysis of CLL samples, we first identified an expression pattern from 17 metabolic-related genes significantly altered through disease progression (i.e., Binet stages A to C). These genes include: Hypoxia-inducible factor-alpha (HIF-A); ATP citrate lyase (ACLY); Long-chain fatty acyl-CoA ligase (ACSL1, 3 and 4); Pyruvate dehydrogenase (PDH); Fatty acid synthase (FASN); Acetyl-CoA carboxylase (ACACA); Malonyl-CoA decarboxylase (MCD); Isocitrate dehydrogenase (IDH); Fumarate hydratase (FH); Aspartate Beta-Hydroxylase (ASPH); Alpha-ketoglutarate-dependent dioxygenase (FTO); Myeloid cell leukemia-1 (MCL1); Malonyl-CoA Decarboxylase (MLYCD); Pyruvate dehydrogenase phosphatase (PDP1); and, Protein Kinase AMP-Activated Catalytic Subunit Alpha 1 (PRKAA1) ( Figure 4B ). These genes were then considered as metabolic reprogramming indicators for future experiments. In line with these findings, we also observed that expression levels of Transcription factor A mitochondrial (TFAM) and the Nuclear respiratory factor 1 (NRF1), two prominent factors involved in mitochondrial biogenesis, were concomitantly increased with CLL staging ( Figure 4C ). These results not only demonstrate that CLL progression directly correlates with rising mitochondrial content in cells; but also, with the levels of metabolic-gene expression.

Using our enrichment analysis, we then categorically sorted our lists of modulated gene expressions, along with their associated pathways, to reveal the most active signaling cascades upon the onset of CLL development (Binet stage A). For example, NF-κB and survival signals (R-HSA-209560); PKMTs methylate histone lysines (R-HSA-3214841); and Interleukin-1 processing (R-HSA-448706) were all highly activated in early stages of CLL disease ( Figure 4D ). On the other hand, signaling pathways associated with advanced disease stages (Binet stage C) identified the pathways of ChREBP activates metabolic gene expression (R-HSA-163765), TNFR1-induced proapoptotic signaling (R-HSA-5357786), and Fatty Acyl-CoA Biosynthesis (R-HSA-75105), all which mitochondria plays a central role in advanced CLL disease (Binet stage C) ( Figure 4E ). These findings indicate that metabolic gene expression supporting metabolic reprogramming are tightly correlated to CLL progression and suggest that these gene profiles are likely impactful regulators of this process.

Based on our results showing that PMPs enable CLL metabolic reprogramming, we investigated the potential impact of PMPs on metabolic transition as part of CLL development and progression. For this purpose, CLL cells were co-cultured with PMPs (1:100) and tested for changes in metabolic-genes expression (identified in Figure 4 ) associated with disease progression. These experiments were also combined with treatments of common anti-leukemia drugs (i.e., Cytarabine, Venetoclax, and Plumbagin) as these compounds elicit well-known mitochondrial inhibitory features. First, we established the median lethal dose (LD50) of anti-cancer drugs for both CII and MEC-1 cell lines as follows: ~100 nM and ~150 nM, respectively, for Cytarabine; ~50 nM and ~75 nM, respectively, for Venetoclax; and ~50 nM and ~ 75 nM, respectively, for Plumbagin ( Figures 5A, B ). The correlation between cell viability and metabolic-reprogramming in PMP-recipient CLL cells was then investigated using RT-qPCR to measure the relative expression levels from our panel of metabolic gene. We observed that PMP treatment of CLL cells induced the expression levels from all metabolic genes associated with advanced disease progression in both CII ( Figure 5C ) and MEC-1 ( Figure 5D ) cells with or without anti-cancer drug (LD50) treatments. For example, PMP treatment induced the expression of HIF-1A, which is known to promote the interaction between neoplastic B cells and their microenvironments (73), induces the generation of tumor-associated macrophages (TAM) (74), and participate in drug-resistance (75). HIF-1A expression was increased 4.3-fold in PMP-recipient CII cells compared to non-treated cells alone. However, this difference was increased in the presence of Cytarabine (6.9-fold), Venetoclax (7.4-fold), and Plumbagin (6.7-fold). This data demonstrates that PMPs increase the expression profiles of metabolic genes associated with CLL progression.

We further repeated this assay with CLLs co-incubated with (1:100) isolated Set2-Mitos in order to evaluate the role of mitochondria acquisition in the observed metabolic reprogramming ( Supplementary Figure 5 ). Similar to PMP-recipient cell groups, our observations indicated a statistically significant improvement in metabolic gene expression patterns in response to Set2 mitochondria internalization into the CLL cells. Meanwhile, this expression pattern didn’t markedly change in the presence of the determined LD50 concentration of Cytarabine, Venetoclax, and Plumbagin for CII and MEC-1 cells. This data demonstrates that the observed alteration in metabolic gene expression patterns of CLLs co-incubated with PMPs mostly derived from mitochondrial content increment.

Reports have previously characterized CLL cells with substantially greater levels of mitochondrial mass and function when compared to healthy B-cells (11, 15). Given our observations of PMP-induced expression of metabolic genes associated with CLL disease progression, we set out to examine if this differential display in metabolic gene expression translates into altered cancer cell phenotypes associated with CLL progression. We therefore co-cultured CLL cells with PMPs (1:100 for 48 hours) and examined cell viability, growth, and programmed cell death (apoptosis) in response to anti-cancer drug treatments (LD50 for 24 hours) using Annexin-V/PI labeling. The stained cells were then analyzed by flow cytometer and sorted into four distinctive quarter zones (Q) including necrotic cells or Q1 (Anx^-/PI⁺), late apoptotic or Q2 (Anx⁺/PI⁺), early apoptotic cells or Q3 (Anx⁺/PI^-), and viable cells or Q4 (Anx^-/PI^-).

As expected, anti-cancer drug treatments induced apoptosis in both CLL models. For example, drug treatments induced early apoptosis (Q3) to 30.3% (Cytarabine), 43.5% (Venetoclax), and 48.2% (Plumbagin) in CII cells ( Figure 6A ). Meanwhile, drug treatments in MEC-1 cells induced early apoptotic events (Q3) by 41.6% (Cytarabine), 38.9% (Venetoclax), and 41.4% (Plumbagin) under the same conditions ( Figure 6B ). However, upon co-culturing conditions with PMPs (1:100, 48 hours), cells were completely refractory to drug toxicity. For example, CII cells treated with PMPs underwent early apoptosis (Q3) in only 2.2% (Cytarabine), 5.65% (Venetoclax), and 4.33% (Plumbagin) in comparison to drug-treated control cells ( Figure 6A ). This represents an inhibition of approximately 93.2% (Cytarabine), 87.1% (Venetoclax), and 91.1% (Plumbagin) in comparison to drug-treated control cells (without PMPs). Similar results were also obtained in MEC-1 co-cultured with PMPs where early apoptosis was inhibited by 97.8% (Cytarabine), 96.8% (Venetoclax), and 89.1% (Plumbagin) in comparison to drug-treated controls ( Figure 6B ). PMP treatments of CLL cells also noticeably increased the viable (Q4) population by 1.89-fold (Cytarabine), 1.85-fold (Venetoclax), and 1.74-fold (Plumbagin) when compared to the non-PMP-treated control cells ( Figure 6A ). Likewise, MEC-1 co-cultures with PMPs increased viability by 1.83-fold (Cytarabine), 1.85-fold (Venetoclax), and 1.87-fold (Plumbagin) when compared to drug treatments in cells alone ( Figure 6B ). These results demonstrate that PMP-treated CLL cells are more resistant to conventional anti-cancer drug toxicity.

To have a better conclusion, this test was also repeated with CLLs co-incubated with (1:100) isolated Set2-Mitos ( Supplementary Figure 6 ). A similar observation has been achieved through flow cytometry assessment of cell proportions in apoptosis when leukemic cells received isolated mitochondria prior to drug treatment. Collectively, this data indicates that mitochondria acquisition in CLL cells has a major impact on cancer cells performance against the therapeutic agents.

To provide insight into the regulation of CLL cycling processes modulated by PMPs, we performed cell-cycle distribution analysis on CLL models treated (or not) with anti-cancer drugs. Our results first show that the basal proportion of the PMP-recipient CLL cells in G2/M (dividing and growing cells) phase was about 1.27-fold in CII ( Figure 7A ) and 1.2-fold higher in MEC-1 ( Figure 7B ) when compared to non-treated control cells. These observations demonstrate that PMPs promote CLL cell cycling. Cell cycle analysis also confirmed our previous observations where toxic drug treatments of CLL cells led to an accumulation of cells in sub-G1 (<G0/G1), a phase representing cellular debris, late apoptotic, and necrotic cells. However, PMP treatments of CLL cells completely inhibited sub-G1 (<G0/G1) accumulation of both CII ( Figure 7A ) and MEC-1 ( Figure 7B ) cell models treated with anti-cancer drugs.

Furthermore, when CLL cells were co-cultured with PMPs prior to anti-cancer drug treatments, cell accumulation in G2/M was increased 10.1-fold (Cytarabine), 7.5-fold (Venetoclax), and 4.0-fold (Plumbagin) for CII cells ( Figure 7A ). Similar observations were noted in MEC-1 cells ( Figure 7B ) where cell accumulation in G2/M phase was enhanced 4.3-fold (Cytarabine), 5.4-fold (Venetoclax), and 4.4-fold (Plumbagin) when treated with PMPs). These results demonstrate that PMPs promote CLL growth and provide anticancer drug resistance.

In addition to the latter results, cell cycle analysis of CLLs incubated with mitochondria (1:100) isolated from Set2 cells also showed a statistically decrement in cells number accumulated in >G0/G1 following treatment with the LD50 dose of Cytarabine, Venetoclax, and Plumbagin chemo-drugs ( Supplementary Figure 7 ). The data supports the notion that mitochondria acquisition via PMPs in CLL cells lowers sensitivity to standard leukemic treatments.

Given our observations that PMPs induce CLL aggressivity through enhanced growth and drug resistance, we wanted to determine the impact of PMPs on other CLL malignant processes such as migration and invasion properties. First, migration assays were performed and indicated that the mobility of leukemic cells treated with PMPs (1:100, 48 hours) was significantly increased when compared control samples without PMPs. Specifically, relative migration of CLL co-cultures was increased by 11.1- and 9.1-fold in CII and MEC-1, respectively ( Figure 8A ). We next evaluated migration properties in the presence of anti-cancer drug treatments. We found that the migration capacity of CLL cells co-cultured with PMPs was significantly improved following drug treatments with Cytarabine (4.5-fold), Venetoclax (6.2-fold), and Plumbagin (5.0-fold) ( Figure 8A ). Similar results were also obtained from MEC-1 cells where PMPs increased migration by 5.2-fold (Cytarabine), 5.5-fold (Venetoclax), and 4.0-fold (Plumbagin) when compared to drug treated control cells ( Figure 8A ).

To evaluate invasion potential, we made use of cell culture inserts layered with Matrigel^® mixture. We found that basal invasion capacity of CLL cells co-culture with PMPs was significantly increased by 11.6- and 10.9-fold in CII and MEC-1 cells, respectively when compared to non-treated CLL cells ( Figure 8B ). These trends were also maintained following drug treatments where PMPs significantly increased the invasion capacity 4.3-fold (Cytarabine), 4.1-fold (Venetoclax), and 6.4-fold (Plumbagin) of CII cells. Equally, MEC-1 cells co-cultured with PMPs also exhibited increased invasion capacity following drug treatments about 5.2-fold (Cytarabine), 4.0-fold (Venetoclax), and 6.0-fold (Plumbagin) higher than their drug-treated control cells ( Figure 8B ). Together, these results demonstrate that PMPs promote CLL invasive features (i.e., migration and invasion) even in the presence of anti-cancer drugs.

To evaluate the role of mitochondria in these events, we performed migration/invasion analysis on CLLs incubated with Set2-derived mitochondria (1:100). We observed a significant increase of mobility ( Supplementary Figure 8A ) and invasiveness ( Supplementary Figure 8B ) following mitochondria acquisition even after treating with an LD50 dose of chemo-drugs including Cytarabine, Venetoclax, and Plumbagin in comparison to control cells. These observations highlight the importance of mitochondrial function in cancer cell resistance to anticancer drugs. Together, these results strongly suggest an impactful role for PMPs in enhanced CLL malignancy, anti-cancer drug resistance, and disease progression.