Isolation, Characterization, and Therapeutic Application of Extracellular Vesicles from Cultured Human Mesenchymal Stem Cells

Stem cells are undifferentiated pluripotent cells with self-renewal capability and translational potential¹. Mesenchymal stem cells (MSCs) are easily isolated, cultured, expanded, and purified in the laboratory, which remains characteristic of stem cells after multiple passages. In recent years, increasing evidence has supported the view that MSCs act in a paracrine mode in therapeutic use^2,3. Especially the secretion of extracellular vesicles (EVs) plays a crucial role in mediating the biological functions of MSCs. As heterogeneous membranous nanoparticles released from most cell types, EVs consist of subcategories termed exosomes (Exos), microvesicles (MVs), and even larger apoptotic bodies^4,5. Among them, Exos is the most widely studied EV with a size of 40-150 nm, which is of an endosomal origin and actively secreted in physiological conditions. MVs are formed by shedding directly from the surface of the cell plasma membrane with a diameter of 100-1,000 nm, which are characterized by high expression of phosphatidylserine and expression of surface markers of donor cells⁶. EVs contain RNA, proteins, and other bioactive molecules, which have similar functions to the parent cells and play a significant role in cell communication, immune response, and tissue damage repair⁷. MSC-EVs have been widely investigated as a powerful cell-free therapeutic tool in regenerative medicine⁸.

Isolation and purification of MSC-derived EVs is a common issue in the field of research and application. At present, differential and density gradient ultracentrifugation⁹, ultrafiltration process¹⁰, immunomagnetic separation¹¹, molecular exclusion chromatograph¹², and microfluidic chip¹³ are widely employed methods in the isolation and purification of EVs. With the advantages and disadvantages of each approach, the quantity, purity, and activity of collected EVs cannot be satisfied at the same time^14,15. In the present study, the differential centrifugation protocol of isolation and characterization of EVs from cultured MSCs is shown in detail, which has supported efficient therapeutic use^{16,17,18,19,20}. The adaptability of this method for a series of downstream approaches, such as fluorescent labeling, local transplantation, and systemic injection, has further been exampled. Implementing this procedure will address the need for simple and reliable collection and application of MSC-EVs in translational research.

All animal procedures were approved by the Animal Care and Use Committee of the Fourth Military Medical University and performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Eight-week-old C57Bl/6 mice (no preference for either females or males) were used. Cryopreserved human umbilical cord-derived MSCs (UCMSCs), used for the present study, were obtained from a commercial source (see Table of Materials). The use of human cells was approved by the Ethics Committee of Fourth Military Medical University.

1. Culture of human mesenchymal stem cells (hMSCs)

1. Prepare the accessories and solutions required for the experimental work, including pipettes, pipette tips, 10 cm Petri dish, thermostatic equipment (water bath), rubber gloves, 75% alcohol, culture medium (incubated at 37 °C), fatal bovine serum (FBS), and 100x penicillin-streptomycin (P/S) solution (see Table of Materials).
2. Prepare commercially available alpha-Minimum Essential Medium (α-MEM) by supplementing with 20% FBS and 1% P/S.
   NOTE: Dilute all the solutions used for isolation and analysis of EVs in ultrapure filtered water made by the ultrapure filtered water purification system (see Table of Materials).
3. Recover the cryopreserved human umbilical cord-derived MSCs (UCMSCs) from the liquid nitrogen by melting them in a 37 °C water bath.
4. Quickly and gently transfer the melted cell suspension into a 15 mL tube with 5 mL of culture medium (step 1.2). Centrifuge at 4 °C, 500 x g for 5 min.
5. Remove the supernatant with a sterile pipette and retain the cell precipitate. Add 2 mL of culture medium into the centrifuge tube and gently resuspend the cells.
6. Seed the cells in a 10 cm Petri dish, and add 8 mL of culture medium to a total of 10 mL.
7. Culture MSCs at 37 °C in 5% CO₂ in the cell incubator.

2. Differential centrifugation for isolation of Exos and MVs

1. When MSCs grow to >90% confluence, replace the culture medium with α-MEM supplemented with 20% EV-depleted FBS and 1% P/S for 8 mL of each dish.
   NOTE: Centrifuge FBS at 4 °C, 1,50,000 x g overnight to remove EVs and other impurities.
2. After 48 h of culture, collect the culture medium (step 2.1) into 50 mL centrifuge tubes.
3. Centrifuge the MSC culture medium at 4 °C, 800 x g for 10 min.
4. Remove the cell fragments and cellular debris by transferring the supernatant to clean 1.5 mL conical tubes.
   NOTE: Using 1.5 mL conical tubes is necessary to increase EV yield.
5. Centrifuge the supernatant at 4 °C, 16,000 x g for 30 min. The pellet is MVs. MVs from each dish of cells were resuspended with 50 µL of PBS.
6. Transfer the supernatant obtained by centrifugation in step 2.5 by pipette to clean ultracentrifuge tubes. Centrifuge at 1,50,000 x g for 2 h at 4 °C.
7. Discard the supernatant and collect the residue, which is Exos. Resuspend Exos from seven dishes of cells in 50 µL of PBS.
   NOTE: The sixth-passaged UCMSCs are used for EV isolation. To get enough MVs and Exos for experimental use, 7-8 dishes of cultured cells are usually needed. The EV samples are used immediately for the following characterization steps or for therapeutic application, which is better not stored.
   CAUTION: Avoid repeated freezing and thawing of EVs, and it is best to store EVs at 4 °C in the short term and not at -80 °C.

3. Detection of particle numbers and size distribution of EVs from MSCs

NOTE: For size distribution evaluation, nanoparticle tracking analysis (NTA) is performed by a commercially available nanoparticle tracking analyzer (see Table of Materials).

1. Dilute 1 µL of EVs (from step 2.5 and step 2.7) in 1,499 µL of PBS and mix enough in a 15 mL tube.
2. Operate the tracking analyzer following the manufacturer's instructions. Firstly, start the compatible software (see Table of Materials).
3. Fill the sample cell with distilled water, and then allow the instrument to start the cell check.
4. Calibrate the instrument with a prepared standard solution. Ensure that the particle number displayed on the software detection interface is between 50-400 (preferably around 200). Click OK.
   NOTE: Prepare the standard solution by reconstituting 1 µL of calibration solution (see Table of Materials) in 1 mL of distilled water to generate a primary solution, and then take 100 µL of the primary solution and add 25 mL of distilled water to prepare a standard solution (1:250,000). Store this working reagent at 4 °C for a week.
5. Rinse the sample cell with 5 mL of distilled water.
6. Before sample analysis, flush the channel with 1 mL of distilled water. Ensure that the particle number displayed on the software detection interface is less than 10.
7. Inject 1 mL of the EV sample prepared in step 3.1. Perform vesicle test according to the operating instructions of the instrument.
   CAUTION: The sample and distilled water are injected with a 1 mL syringe at a constant speed of 0.5 mL/s under careful manual control.

4. Characterization of EV morphology by transmission electron microscope (TEM)

1. Dilute 1 µL of EVs from step 2.7 in 199 µL of PBS and mix enough. Add 20 µL of EV suspension drops to the formvar/carbon-coated square mesh (see Table of Materials), and let it stay for 3 min.
2. Remove excess liquid by a small filter paper, and let it stay for 15 s to slightly dry the surface.
3. Negatively stain EV samples with 1.5% phosphotungstic acid droplets for 40 s.
4. Remove the excess phosphotungstic acid and allow it to stand for 15 s to slightly dry the surface.
5. Put the sample containing formvar/carbon-coated square mesh into a clean dish covered with filter papers. Observe and capture images with a transmission electron microscope.
   NOTE: In this process, Exos are taken as an example.

5. Labeling of MSCs-derived EVs

1. After EVs extraction (step 2.5), resuspend with 250 µL of PBS in a 1.5 mL conical tube.
2. Use another 1.5 mL conical tube for preparing the working solution of the PKH26 dye; use 1 µL of PKH26 labeling reagent diluted with 250 µL of Diluent C (a component of the commercially available EV labeling kit used for the present study, see Table of Materials).
   NOTE: Prepare the working solution immediately before use.
   CAUTION: Excessive PKH26 dye or labeling without pre-dilution is at risk of damaging the EVs.
3. Mix the resuspended EVs with the working solution. Allow it to stand at room temperature for 5 min, and then add 500 µL of EV-depleted FBS to stop the reaction.
4. For MVs, centrifuge at 4 °C, 16,000 x g for 30 min, and discard the supernatant by a pipette. Add 1 mL of PBS to rinse the residue.
5. Centrifuge at 4 °C, 16,000 x g for 30 min, and discard the supernatant to remove the unbound dye.
6. Resuspend the residue with 200 µL of PBS. Drop 20 µL of suspension on the slide and observe it under a fluorescence microscope.
   NOTE: In this process, MVs are taken as an example.

6. Local transplantation and systemic injection of MSC-EVs

NOTE: For the following procedures, place the EVs on ice before injection.

1. Perform local transplantation following the steps below.
   1. Anesthetize the mice with 4% (vol/vol) isoflurane. Maintain anesthesia at 1%-3% isoflurane during the procedure.
   2. Before the surgery, administer 5 mg/kg carprofen (IP) for analgesia to minimize postprocedural pain.
   3. Before wound excision, shave the dorsal hair and sterilize the dorsal surface by applying 3 alternating rounds of 10% povidone-iodine and 75% ethanol. 
   4. Using a biopsy punch (see Table of Materials), create a full-thickness wound of 1 cm in diameter on the dorsal skin.
   5. Resuspend the MSC-derived EVs from two dishes in 100 µL of PBS and administer the injection in the subcutaneous layer of the wound bed, around each wound with four sites.
      NOTE: An average of 100 µg of EVs per mouse (EVs from two 10 cm dishes of cells) is injected. Ten 10 cm dishes of MSCs are needed to yield enough EVs to set up an experiment of n = 5 in a murine model.
   6. After treatment, wrap the mice with sterile gauze and place them on heat insulation pads (see Table of Materials) until they wake up.
   7. Ensure that the mice are not left unattended until they have regained sufficient consciousness. Do not return the mice to the company of other animals until they have fully recovered.
   8. Administer additional doses of analgesia on day 2 and day 3 post-surgery as required. 
2. Perform systemic injection.
   1. Resuspend MSC-derived EVs in 200 µL of PBS, and then mix with heparin solution at a 10:1 volume ratio.
   2. Place the mice into the caudal vein imaging system (see Table of Materials). Press the switch to lift the lever.
   3. Disinfect the tail vein using an alcohol pad before the IV injection. Then, inject the mice systematically with suspension prepared in step 6.2.1 through the tail vein. The procedure is helped by a tail vein illustrator.
      NOTE: Inject EV suspension immediately after mixing with heparin. An average of 100 µg EVs per mouse (EVs from two 10 cm dishes of cells) is injected. Ten 10 cm dishes of MSCs are needed to yield enough EVs to set up an experiment of n = 5 in a murine model.
      CAUTION: Mouse tail vein injection is usually limited to 200 µl of volume. The injection should go slowly and stop if you see a bump or resistance, which suggests the needle is out of the vein. If the mice show adverse clinical signs like curling and trembling, this could be a shock response to high-volume injection, fast injection, or toxicity/anaphylaxis.

MVs and Exos from cultured human UCMSCs are isolated following the experimental workflow (Figure 1). The NTA results demonstrate that the size of Exos from human MSCs ranges from 40 nm to 335 nm with a peak size of about 100 nm, and the size of MVs ranges from 50 nm to 445 nm with a peak size of 150 nm (Figure 2). Morphological characterization of MSC-derived Exos exhibit a typical cup shape (Figure 3). EVs are efficiently labeled by PKH26, which is observed both by the gross view as labeled pellets and by fluorescent microscopy (Figure 4). Local and systemic injections of MSC-derived EVs are performed around the skin wound or through the caudal vein (Figure 5). Thus, the Exos and MVs are successfully isolated and applied for subsequent experiments.

Figure 1: Isolation of MVs and Exos from cultured human MSCs. Culture media is collected and differential centrifugation is performed. Please click here to view a larger version of this figure.

Figure 2: NTA analysis of Exos and MVs from MSCs. Particle size distribution with representative images is depicted. Please click here to view a larger version of this figure.

Figure 3: Morphology characterization of Exos from MSCs by TEM. Cup-shaped Exos are displayed. Scale bars: 200 nm. Please click here to view a larger version of this figure.

Figure 4: Labeling of MSC-derived MVs by PKH26. PKH26-labeled MV pellets are grossly observed and viewed under a fluorescent microscope. Scale bar: 200 µm. Please click here to view a larger version of this figure.

Figure 5: Administration of MSC-derived EVs. (A) Local transplantation around the skin wound and (B) Systemic injection via the caudal vein. Please click here to view a larger version of this figure.

EVs are emerging to play an important role in diverse biological activities, including antigen presentation, genetic material transport, cell microenvironment modification, and others. Furthermore, their wide application brings new approaches and opportunities for diagnosing and treating diseases²¹. Implementation of therapeutic applications of EVs is based on successful isolation and characterization. However, due to the lack of standardized isolation and purification methods and the low extraction efficiency, EV research has been hampered. This article mainly introduces an easily feasible extraction and characterization protocol of EVs from cultured human MSCs and further applications, which can be labeled and used to promote tissue repair via local injection and treat systemic diseases by intravenous injection. Reproducibility of MSCs for EV isolation is preserved by using the neonatal origin of UCMSCs, standard culture of cells, and no late passages. Although the term Exos and MVs are respectively used in this manuscript to refer to EVs collected from differential centrifugation speeds, more assays will be needed in the future studies to distinguish their origins, as recommended by the Minimal Information for Studies of Extracellular Vesicles (MISEV2018)²² guideline, such as by live-cell imaging for observing the production process and by immuno-electron microscope for detecting the specific markers. Overall, the method is simple and thus readily scalable and reproducible, which will be useful to the field.

In the present study, physiological EVs from MSCs are progressively separated into MVs and Exos by differential centrifugation and ultracentrifugation. Here, the significance of replacing α-MEM supplemented with 20% EV-depleted FBS for collecting the culture medium is emphasized. This step ensures that the extracted EVs are derived only from MSCs, excluding the possibility that EVs might be from other interfering substances. It is also critical to mix 10% heparin solution into the EV solution before systemic injection. In recent years, several studies have shown that EVs have a procoagulant effect, so adding heparin during systemic injection is necessary to avoid the formation of thrombus after injection, which otherwise leads to acute lethality of mice²³. There might be some issues during this protocol implementation that need troubleshooting. For one prominent example, if the extraction quantity of EVs is not enough, researchers should trace it back to the cell supernatant collection step. During supernatant collection, cells must be blown evenly to fully collect the released EVs retained in the extracellular matrix of MSCs. Another potential failure occurs when mice die soon after the injection of EVs. Besides adding heparin before injection, as mentioned above, the concentration of injected EVs (an average of 100 µg EVs per mouse) must be adjusted. To verify the success of the injection, in vivo imaging of biodistribution of fluorescence-labeled EVs or frozen sections of specific tissue sites of engraftment (the liver, the spleen, etc.) can be performed, which was previously reported²⁴. Also, EVs are recommended to be used as soon as possible after isolation, and freezing storage will cause particle loss, purity reduction, and fusion phenomena leading to artefactual particles²⁵. Besides, the commonly adopted PKH26 labeling will increase the EV size²⁶, and other labeling methods can be used for comparison.

The existing protocols for isolating EVs are mainly reported as follows: differential and density gradient ultracentrifugation⁹, ultrafiltration process¹⁰, immunomagnetic separation¹¹, molecular exclusion chromatograph¹², microfluidic chip¹³, and polymer-based precipitation technology²⁷. Compared to ultracentrifugation, the ultrafiltration process is a convenient extraction method and conducive to the enrichment of EVs in larger volume samples^9,10, while the disadvantage is that there is membrane pollution, which is easy to cause sample loss. Immunomagnetic separation is suitable for separating different subtypes of EVs, and the obtained EVs have high purity. The disadvantage of this method is the high cost with low yield¹¹. The EVs separated by molecular exclusion chromatograph have high purity and high yield. The disadvantage is that special equipment is required and cannot be used for multiple samples at the same time¹². Microfluidic chips have high requirements for materials and technology, with high cost, and it is difficult to process large samples¹³. The EVs separated by polymer-based precipitation technology contain non-EV contaminants that will affect EV activity¹⁴. Among them, differential centrifugation and ultracentrifugation is still the most widely used method for separation and purification of EVs and is regarded as the gold standard for MSC-EV extraction. Separation and enrichment of EVs from MSC culture media are achieved step by step through the combination of different time and speed of centrifugation, which represents an optimized method. It is notable that Exos and MVs isolated in this manuscript have similar size distributions, although MVs tend to be larger, which might be due to differential centrifugation that tends to induce aggregation of nanoparticles. Although modified methods may perform better in the size maintenance of EVs, differential centrifugation is advantageous in high efficiency for EV separation²⁸. Although for large-scale production of EVs and GMP production, this method is limited by the amount of media that can be processed in one isolation run, the technical system is easy to set up and has been applied by translational projects on large animals^29,30. Anyway, the above separation methods have their advantages and disadvantages, and researchers may choose the appropriate method according to the different sources of EVs and the research aim.

In recent years, EVs have shown great potential in clinical applications, such as disease diagnosis³¹, treatment, and as carriers of drug delivery^32,33. Based on their advantageous properties, various countries have widely studied EVs as therapeutic agents. In this regard, studies have been committed to EVs in disease treatment and the mechanism underlying the role of EVs in the pathogenesis of the disease. The current protocol of isolation, characterization, and therapeutic application of EVs from cultured MSCs has been well applied. For instance, local delivery of MSC-Exos in skin wounds promotes healing by inflammatory modulation and tail-vein injection of MSC-derived EVs can improve hepatic insulin resistance and steatosis in type 2 diabetes mellitus²⁴. Furthermore, therapeutic applications involving increased yield and purity of MSC-EVs are on the horizon to provide feasible translational strategies.