Collection of human skin tissue was approved by Western IRB (Protocol#20203104). All tissue pieces were digested by 3% collagenase type 3 (Worthington Biochemical Corporation, Lakewood, NJ, USA) at 37˚C for 5 hours in a sterile stirring chamber. The liquid was transferred into a sterile 50 ml centrifuge tube and centrifuged for 5 minutes at 400 g. Supernatant was removed. Cell pellet was resuspended in 30 ml pre-warmed (37˚C) complete alpha minimum essential media (alpha MEM) (GIBCO, Carlsbad, CA, USA) containing 5% human platelet lysate (HPL) (Biological Industries, Cromwell, CT, USA) and 0.05 mg/ml gentamicin (GIBCO, Carlsbad, CA, USA). Cells was cultured at 37˚C in 5% CO₂ incubator overnight. The medium and floating cells was aspirated next day. 30ml fresh medium was added to the flask. The flask was returned to the incubator for culture with medium refresh every 3 - 4 days. When cell growth was 80% confluent, cells were sub-cultured to new flasks by 1:10 dilution. The primary fibroblasts of second and third passages were collected, re-suspended in Nutrifreez D10 cryopreservation medium (Biological Industries, Cromwell, CT, USA), aliquoted to 1 ml containing 1 × 10⁶ cells in cryovials, frozen and stored in liquid nitrogen.

Human primary fibroblasts were established and stored in liquid nitrogen [14] . Human primary fibroblasts were recovered from liquid nitrogen and cultured in alpha MEM (GIBCO, Carlsbad, CA, USA) containing 5% human platelet lysate (HPL) (Mill Creek Life Sciences, Rochester, MN, USA). When cells grew to 80% full in flask, they were digested with TrypLE expression solution (GIBCO, Carlsbad, CA, USA) and sub-cultured at 37˚C and 5% CO₂.

BM-MSCs were cultured on laminin-521 coated plates in StemMACS MSC Expansion Medium (Miltenyi Biotec, Germany) at 37˚C and 5% CO₂. Laminin-521 was purchased from Biolamina Ab, Sweden and the cell culture plates were coated according to the manufacturer’s instructions. For production of EVs, BM-MSCs were cultured until approximately 80% confluency and carefully washed twice with PBS. After that, serum-free Opti-MEM^TM medium (GIBCO, Carlsbad, CA, USA) was added and the cells were incubated for 48 hours at 37˚C and 5% CO₂. Then, the conditioned medium was collected, and centrifuged first for 5 minutes at 700 g to remove living cells and a second spin for 100 minutes at 250 g to remove cellular debris. The cells that remain are cultured for 24 hours in the complete culturing medium. Another production round was made as described above.

The centrifuged conditioned medium was sterile filtered and concentrated using tangential flow filtration (TFF), KR2i TFF system (SpectrumLabs). For that, a modified polyethersulfone (mPES) hollow fiber filter with 300 kDa membrane pore size was used (MidiKros, 370 cm² surface area, SpectrumLabs). The flow rate was set at 100 ml/min. At first, the conditioned medium was concentrated to a volume of 75 ml in the container (100 ml total volume in the system) and, then, dialyzed with 1 liter of sterile filtered PBS. After the dialysis, the sample was concentrated to 30 ml and sterilized using a 0.22 µm filter. Finally, the sample was concentrated to a volume of 1 - 2 ml using a 10 ml 10 kDa MWCO spin filter (Amicon).

The samples were mixed with Laemmli Sample Buffer (Bio-Rad) under reducing conditions and warmed at 95˚ for 10 min. Then, 4% - 12% gradient gels were used for SDS electrophoresis and the proteins were transferred to polyvinylidene fluoride membranes using iBlot 2 (Invitrogen) according to the manufacturer’s instructions. The membranes were first hybridized with the antibody of interest and, then, with the corresponding HRP-conjugated secondary antibodies. For visualization, chemoluminescent HRP-substrate from Amersham Biosciences was used and the figures were made using ChemiDoc imaging system (Bio-Rad).

A 5 µl drop of the sample was placed on a formvar and carbon coated 200-mesh copper grid. The excess solution was removed by blotting with filter paper. The sample was then directly contrasted with 2% uranyl acetate. Excess of uranyl acetate was removed by blotting on filter paper. The contrasting step was repeated twice. Dried grids were examined by Tecnai^TM G2 Spirit BioTwin transmission electron microscope (Thermo Fisher/FEI) at 80 kV with an ORIUS SC200 CCD camera and Gatan Digital Micrograph software (both from Gatan Inc.).

Mitochondrially encoded cytochrome c oxidase I (MT-CO1) and mitochondrially encoded NADH dehydrogenase 1 (MT-ND1) in the samples of imEVs were detected by PCR [14] . PCR was performed in a total 25 µl volume including 2 µl of PCR Master mix (Thermofisher Scientific, Waltham, MA, USA), 0.5 µl of 100 µM forward and reverse primers and 1 µl of heat-denatured imEVs samples. The primers for MT-CO1 amplification were 5’-CTAGCAGGTGTCTCCTCTATCT-3’ and 5’-GCTCGTGTGTCTACGTCTATTC-3’. The primers for MT-ND1 were 5’-CCTTCGCTGACGCCATAAA-3’ and 5’-GGTCTCTGCTAGTGTGGAGATA-3’. β-actin (ACTB) was used as endogenous gene control. The primers for ACTB amplification were 5’-CCTTTCTCACTGGTTCTCTCTTC-3’ and 5’-CGTAGCACAGCTTCTCCTTAAT-3’. The amplification procedure included initial DNA denaturing at 95˚C for 3 minutes, then 35 cycles of denaturing 30 seconds at 95˚C, primer annealing 30 seconds at 50˚C and 60 seconds of extension at 72˚C, and final extension of 10 minutes at 72˚C in a T100 Thermal Cycle (Bio-Rad, Hercules, CA, USA). PCR products were run on a 2% agarose gel and imaged by ethidium bromide fluorescence.

Mitochondrial isolation follows a previously described protocol [14] . All reagents were sterile. Fibroblasts were centrifuged for 5 minutes at 400 g and at 4˚C to remove the media. Cell pellet was re-suspended in ice-cold 300 mM sucrose mitochondrial isolation buffer (MIB) (Sigma Aldrich, St. Louis, MO, USA) and homogenized by bead beating (Bead Ruptor 12, Omni International homogenizer company, Kennesaw, GA, USA). The cell lysate was centrifuged for 10 minutes at 700 g and at 4˚C. Then the supernatant was transferred to new centrifugation tubes and centrifuged for 10 minutes at 9000 g at 4˚C. The supernatant was removed. The wet weight of mitochondria was measured. The mitochondrial pellet was re-suspended with 240 mM sucrose mitochondrial respiration buffer (MRB) (Sigma Aldrich, St. Louis, MO, USA). We mixed the isolated mitochondria of fibroblasts with imEVs (9:1 in volume) and stored the mixture at 2˚C - 6˚C for 2 and 5 days. MMP and mitochondrial ATP content were measured and compared to them of fresh mitochondria (0 day).

MMP generated by proton pumps is an essential component in the process of energy storage during OXPHOS. Membrane potential dependent dyes such as JC-1 (5,5’,6,6’-tetrachloro-1,1’,3,3’-tetraethylbenzimidazolocarbocyanine iodide) and MitoTracker dyes (rosamine- or cyCarbocyanine-based probes) have been used to stain mitochondria and monitor mitochondrial potential [15] . The stock solution of JC-1 is added to mitochondrial suspension to a final concentration 1 µg/ml. The mixture is incubated for 10 minutes at room temperature. For a control assay, stock solution of valinomycin is added to mitochondrial sample to a final concentration of 10 µg/ml. The mitochondrial sample containing valinomycin is kept on ice for 10 minutes to allow complete dissipation of the MMP. The valinomycin-treated mitochondria are stained with JC-1 dye for 10 minutes at room temperature. Red fluorescent J-aggregates in intact mitochondria can be observed under fluorescent microscope. The relative fluorescence units (RFU) can be read in multiple plate fluorimeter using end-point method with the setting of Ex/Em: 490 nm/590nm.

ATP content is measured with ATPlite kit (Perkin Elmer Inc., Waltham, MA, USA). The detailed procedure is referred to the product manual. The brief method is as follows: add 50 µl of mammalian cell lysis solution to 100 µl of mitochondria, the mixture of mitochondria with imEVs, or MRB per well in a 96-well plate with white well and clear bottom; shake the plate for 5 minutes; add 50 µl substrate solution to all wells and shake the plate for 5 minutes; measure the luminescence of the plate; calculate ATP content of samples using the ATP standard curve.

Student’s t-test was used to test statistical significance. p-value less than 0.05 was judged to be of statistical significance.

imEVs were characterized according to the International Society for Extracellular Vesicles guidelines. NanoSight 3000 device and transmission electron microscopy revealed the characteristic sizes and shapes for exosomes. The EV samples expressed positive markers of EVs (CD63, CD81, and TSG101) but negative marker of EVs (Calnexin) by using West-blotting.

imEVs were shipped on dry ice and stored at −20˚C after arrival. At 7 days post-freezing, the imEVs were thawed at 4˚C and stained with JC-1. The mitochondrial genes in the imEVs were determined by PCR. We found intact mitochondria that actively concentrated JC-1 dye in the imEVs sample (Figure 1). Mitochondrial genes of MT-ND1 and MT-CO1 in the imEVs were amplified by PCR (Figure 2). The JC-1 staining and PCR confirmed the imEVs contained viable mitochondria. The viable mitochondria maintained MMP even though the imEVs went through freezing-thawing cycles. The viable mitochondria may be apoptotic bodies or free mitochondria in the imEVs solution. The results suggest that imEVs may enhance the stability of extracellular mitochondria.

The mitochondria and the mixture of mitochondria with imEVs were stored at

2˚C - 6˚C for 2 and 5 days. Microscopy showed that the imEVs stimulated aggregation of the isolated mitochondria and maintained MMP, compared to the mitochondria without imEVs (Figure 3). MMP was measured by a multiple plate fluorimeter using end-point method with the setting of Ex/Em: 490 nm/590nm. The MMP was compared to the MMP of fresh mitochondria (at 0 day). The isolated mitochondria stored at 2˚C - 6˚C for 2 days lost significant MMP (RFU 5458 ± 52 at 0 day vs. 1822 ± 68 at 2 days, p < 0.01). The MMP at 2 days was approximately 33% of MMP at 0 day. However, the mitochondria mixed with imEVs maintained and even slightly increased the MMP after 2 days of storage at 2˚C - 6˚C (RFU 5962 ± 222 at 0 day vs. 6786 ± 291 at 2 days, p > 0.05). Mitochondria with imEVs have significantly higher MMP than the mitochondria without imEVs (6786 ± 291 vs. 1822 ± 68, p < 0.01) after 2 days of preservation at 2˚C - 6˚C (Table 1). After 5 days, both the mitochondrial without and with imEVs lost 73% and 70% of MMP (p < 0.01, P < 0.01, compared to fresh mitochondria), respectively (Table 1, Figure 4). The preliminary results showed that imEVs prolonged the survival of isolated mitochondria in cold storage at least 2 days.

¹Relative fluorescence units (RFU), mean ± standard deviation (SD) (N); ²Compared to fresh mitochondria (0 day); ³Compared to the Mitochondria (without imEVs).

In agreement with the MMP, the isolated mitochondria lost approximately 40% ATP content after 2 days cold storage at 2˚C - 6˚C, compared to fresh isolated mitochondria (0 day) (p < 0.01). In contrast, the mitochondria mixed with imEVs maintained approximately 90% ATP content, compared to fresh mitochondria (p > 0.05) (Table 2). Mitochondria with imEVs have significantly higher ATP content than the mitochondria without imEVs (90 ± 8 vs. 60 ± 9, p < 0.05) (Table 2). After 5 days of cold storage, mitochondria in both solutions lost significant ATP content (p < 0.01, p < 0.01, respectively, compared to fresh mitochondria) (Table 2, Figure 5).