Original Article

Cell Transplantation
Volume 30: 1–13
Comparison and Investigation ª The Author(s) 2021
Article reuse guidelines:
sagepub.com/journals-permissions
of Exosomes Derived from Platelet-Rich DOI: 10.1177/09636897211017833
journals.sagepub.com/home/cll
Plasma Activated by Different Agonists

Shunli Rui1, Yi Yuan2, Chenzhen Du2, Peiyang Song2, Yan Chen2, Hongyan Wang2,
Yahan Fan3, David G. Armstrong4, Wuquan Deng2 , and Ling Li1

Abstract
PRP-Exos are nanoscale cup-shaped vesicles that carry a variety of proteins, mRNAs, microRNAs, and other bioactive substances.
PRP-Exos can be formed through several induction pathways, which determine their molecular profiles and facilitate their tai-
lormade participation in intercellular communication. Currently, little is known on how the PRP-Exos activation method influences
the quality and quantity of PRP-Exos. The present study aims to observe and analyze the number, profile, and growth factors of PRP-
Exos through TEM, Nanoflow, and WB after PRP activation and compare the difference in function of PRP-Exos on HUVECs, with
different stimuli (calcium gluconate, thrombin, or both). We found that PRP activated with both thrombin and calcium gluconate
harvested the highest concentration of exosomes [(7.16 + 0.46)  1010 particles/ml], compared to thrombin group [(4.87 + 0.15)
 1010 particles/ml], calcium gluconate group [(5.85 + 0.43)  1010 particles/ml], or saline group [(7.52 + 0.19)  109 particles/
ml], respectively (P < 0.05) via Nanoflow analysis. The WB analysis showed that cytokines (VEGF, PDGFBB, bFGF, TGF-b) are
differentially encapsulated in PRP-Exos, depending on the PRP stimulus, in which the mixture-PRP-Exos yielded the highest con-
centration of cytokines. In the function assay of PRP-Exos on HUVECs, the mixture-PRP-Exos promoted HUVECs proliferation,
increased HUVECs migration, promoted the formation of vessel-like by HUVECs via the AKT ERK signal pathway more dra-
matically, compared with other groups. In summary, our studies showed that PRP activated by the mixture of calcium gluconate and
thrombin harvested the best quality of exosomes which had the top biological functions. This study provides a protocol for selecting
appropriate PRP activators to obtain high-quality exosomes for future applications.

Keywords
exosomes, platelet-rich plasma (PRP), calcium gluconate, thrombin, HUVECS

Introduction blood. PRP releases a large number of autologous growth

Platelet-rich plasma (PRP) is a mixture of highly concen- factors and cytokines after activation1. PRP has been widely
trated platelets obtained from centrifugation of fresh whole used in the field of repair and regeneration, including in

1
The Key Laboratory of Laboratory Medical Diagnostics in the Ministry of Education and Department of Clinical Biochemistry, College of Laboratory
Medicine, Chongqing Medical University, Chongqing, China
2
Department of Endocrinology, Multidisciplinary Diabetic Foot Medical Center, Chongqing Emergency Medical Center, Chongqing University Central
Hospital, Chongqing University, Chongqing, China
3
Department of Blood Transfusion, Southwest Hospital, Chongqing, China
4
Department of Surgery, Keck School of Medicine of the University of Southern California, CA, USA

Submitted: November 26, 2020. Revised: April 16, 2021. Accepted: April 23, 2021.

Corresponding Authors:
Wuquan Deng, Department of Endocrinology, Multidisciplinary Diabetic Foot Medical Center, Chongqing Emergency Medical Center, Chongqing University
Central Hospital, Chongqing University, Chongqing 400014, China.
Email: [email protected]
Ling Li, The Key Laboratory of Laboratory Medical Diagnostics in the Ministry of Education and Department of Clinical Biochemistry, College of Laboratory
Medicine, Chongqing Medical University, Chongqing 400016, China.
Email: [email protected]

Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0
License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further
permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).
2 Cell Transplantation

chronic tendinitis 2, osteoarthritis3, chronic wounds4, and Southwest Hospital, Army Medical University, Chongqing,
plastic surgeries5. A series of our previous studies on PRP China. Volunteers did not take any drugs that would have
treatment has revealed positive effects on wound healing in affected platelet number or function within the 2 weeks
patients with diabetic foot ulcers6–9. The basic cytokines before blood donation. All volunteers provided written
identified in platelets include transforming growth factor-b informed consent following the approval of the Ethical Com-
(TGF-b), platelet-derived growth factor (PDGF), basic fibro- mittee Board of Southwest Hospital. Citrate glucose
blast growth factor (bFGF), vascular endothelial growth fac- (22 g/L sodium citrate, 24.5 g/L glucose monohydrate, and
tor (VEGF), and endothelial cell growth factor. These 8 g/L citric acid monohydrate) was used as an anticoagulant
natural cytokines are present in “normal” biological propor- for blood samples at a ratio of 1:10, and 100 ng/mL of
tions and play important roles in cell proliferation, chemo- prostaglandin E1 (PGE1, Sigma-Aldrich, St. Louis, MO,
taxis, cell differentiation, and angiogenesis10. Despite these USA) was added to prevent platelet activation during isola-
advantages, the limitation of PRP applications is the require- tion. PRP (40 mL) was isolated by a fully-automatic blood
ment for autologous platelets. separator (CS-3000 plus; Baxter International, Inc., Deer-
Exosomes are a kind of extracellular vesicles (EVs), ori- field, IL, USA), which has a leukocyte reduction system to
ginating from intracytoplasmatic multivescicular bodies avoid leukocyte contamination. The concentration of apher-
(MVBs), and are directly released into the extracellular esis platelets in PRP and venous blood were counted by the
space upon the fusion of the MVB membrane with the automatic hematology analyzer (KX-21 N; Sysmex Corpo-
plasma membrane11. Platelet-derived exosomes (PLT-Exos) ration, Kobe, Japan). The prepared PRP was stored at 24 C
are stored in MVBs and a-granules12. After activation, the while shaking or stored at -80 C until ready to use.
outer membrane of MVBs is invaded and released out of the
platelets through extracellular secretion. EVs are a means of
communication between long-distance cells in the body13.
PRP Activation
At present, EVs are mainly divided into three categories: For the present study, each PRP was divided into four ali-
exosomes, microvesicles, and apoptotic bodies14. Given that quots of 10 mL and stimulated in the presence of (A)
exosomes have no obvious adverse effects such as immuno- 0.1 U/mL thrombin (Sigma Aldrich, St. Louis, MO, USA);
genicity or tumorigenicity15, the potential of exosomes in (B) 2 mm/L calcium gluconate solution (Sigma Aldrich,
wound healing has attracted increasing attention. St. Louis, MO, USA); (C) a mixture with an equal volume
In 2014, Torreggiani et al.16 first isolated exosomes from of 2 mm/L calcium gluconate solution and 0.1 U/mL throm-
PRP and demonstrated their potential effects on the prolif- bin; or (D) an equal volume of 0.9% saline. These mixtures
eration and migration of mesenchymal stem cells, which was were incubated with agonist or with saline control for 90 min
the first described role of PRP-Exos in tissue regeneration. at 37 C in a water bath.
Furthermore, Tao et al.17 demonstrated that PRP-Exos effi-
ciently promote the re-epithelialization of chronic cutaneous
wounds via activation of Yes-associated protein (YAP) in a
Histological Examination of PRG
diabetic rat model. Although PRP-Exos have been exten- Platelet-rich gel (PRG) was collected from activated PRP,
sively studied in recent years17,18, few studies have assessed and samples were fixed with 4% paraformaldehyde and
the method of activating PRP to release exosomes. embedded in paraffin. Embedded samples were sectioned
The current standards for preparing PRP are not uniform at 5 mm thickness to perform hematoxylin and eosin
because of consensus regarding PRP preparation techniques (H&E) staining. The staining was performed according to
and the complicated PRP activation methods. In this study, the manufacturer’s protocol (Sigma Aldrich, St. Louis, MO,
we aimed to develop a standard protocol for preparing exo- USA). Finally, the paraffin section was sealed using resi-
somes for clinical treatment or research purposes, with con- nene, and pictures were taken under an Olympus BX51 opti-
sideration given to cytokines being predominantly stored in cal microscope (Olympus America, Inc., Center Valley, PA).
PRP-Exos after PRP activation16,19. Previous studies have
shown that PRP-Exos can be obtained through a four-step
method20, but the steps are cumbersome, the agonists cannot
Exosome Separation
be widely used in clinical practice, and the platelets may be Exosomes were isolated and purified by the gradient ultra-
prematurely activated to release exosomes during the separa- centrifugation method created by Thery et al.21. Briefly, PRP
tion and purification steps. was transferred to a 15 mL sterile polypropylene tube, and
different agonists were added to activate platelets to release
exosomes. All centrifugation steps were carried out at 4 C.
Materials and Methods Activated PRP was sequentially centrifuged at 300  g for
10 min, 1500  g for 15 min, and 3200  g for 20 min to
PRP Preparation remove cellular debris. Then, the supernatant was succes-
Venous blood (400 mL) was gained from three healthy sively centrifuged at 20,000  g at 4 C using an SW26 rotor
adults recruited from the Department of Blood Transfusion, (Beckman Instruments, Inc., Fullerton, CA, USA) for 30 min
Rui et al 3

to separate the exosomes from the MVs. After each centri- values during analysis. The number of particles after the
fugation, the pellet was discarded, and the supernatant was sample was diluted was in the optimal range of
used in the following steps. Then, the supernatant was fil- 4000–14,000. Particle concentration and size distribution
tered through a 0.22-mm filter (Millipore, Burlington, MA, were analyzed using the NanoFCM software (NanoFCM
USA). Exosomes were collected by ultracentrifugation using Profession V1.0) and normalized for the starting volume of
an SW26 rotor (Beckman Instruments, Inc., Fullerton, CA, exosome resuspension and the dilution factor that was nec-
USA) at 100,000  g for 70 min at 4 C, washed in sterile essary for proper NanoFCM reading.
1  PBS (no calcium, no magnesium, and no phenol red;
Gibco; Thermo Fisher Scientific, Waltham, MA, USA), and
pelleted by ultracentrifugation at 100,000  g for 70 min. Cell Cultures
The pellets were carefully resuspended in 200 mL of sterile Human umbilical vein endothelial cells (HUVECs) were
1  PBS containing protease inhibitor cocktail (Thermo obtained from ScienceCell and cultured in endothelial cell
Fisher Scientific, Waltham, MA, USA) and phosphatase medium (ECM, ScienceCell Research Laboratories, Carls-
inhibitor cocktail (Thermo Fisher Scientific, Waltham, bad, CA, USA), supplemented with 10% FBS and EC
MA, USA) and frozen at 80 C for future use. In the study, growth supplement (ECGS, ScienceCell Research Labora-
the supernatant of activated PRP was named PRP-AS. The tories, Carlsbad, CA, USA). All experiments were conducted
exosomes isolated from thrombin-activated PRP were with HUVECs in passages 3-6. Cells were incubated in a
named thrombin-PRP-Exos. The exosomes isolated from humidified atmosphere of 5% CO2 at 37 C.
calcium gluconate-activated PRP were named Ca2þ-PRP-
Exos. The exosomes isolated from saline-activated PRP
were named saline-PRP-Exos. The exosomes isolated from Exosomes Labeling, Internalization, and Confocal
mixture-activated PRP were named mixture-PRP-Exos. Microscopy
Exosomes were labeled using PKH-26 (red, Sigma-Aldrich)
TEM Analysis and added into the HUVECs culture at a concentration of 50
mg/ml. After incubation for 6 hours, HUVECs were washed
Exosome suspensions (10 ml) were fixed by the same volume
of glutaraldehyde (filtered twice through the 0.22-mm filter). twice with diluted PBS. Then cells were fixed with 4% par-
aformaldehyde for 20 min at RT. The nuclei of HUVECs
Then, 10 ml of fixed exosomes were coated on 400 mesh
were stained with 40 ,6-diamidino-2-phenylindole (DAPI).
grids (formvar/carbon-coated, glow-discharged) for 90 s.
The uptake of exosomes was observed by the laser scanning
The grids were negatively stained with uranyl acetate (2%)
confocal microscope (Leica Microsystems, Wetzlar,
for 30 s, and excess uranyl acetate was carefully removed
Germany).
using filter paper. The samples were examined immediately.
The size and shape of the exosomes were observed, and
micrograph images were taken by TEM (HITACHI- Cell Proliferation Assay of HUVECs
H7650, Hitachi, Tokyo, Japan) operating at 100.0 kV with
an AMCT XR80 CCD sensor allowed to dry at room tem- HUVECs suspension was seeded in 96-well plates (5  103
perature (RT). cells per well) and were co-cultured with PRP-Exos for
24, 48, 72, and 96 h, respectively. The cells of each group
were treated accordingly, and three parallel wells were set up
Particle Size Statistics of TEM Images in each group. Cell proliferation was analyzed using a Cell
ImageJ 1.42q software (National Institutes of Health) was used Counting Kit-8 (CCK-8, Beyotime, Jiangsu, China).
to calculate the diameter of all exosomes from the TEM
images. The graphs and data analyses were generated by Origin
2019 (OriginLab Corporation, Northampton, MA, USA).
Transwell Migration Assay of HUVECs
HUVECs were seeded to the upper well of transwells in a
6-well plate with an 8-mm pore size (Corning-Costar, Corn-
Nanoflow Analysis ing, NY, USA) at 5  104cells per well, and PBS or PRP-
Exosome samples (10 ml) were diluted (1:100) and analyzed Exos was added to the medium containing 5% serum of the
using the Flow Nano Analyzer (NanoFCM Inc. China), lower chamber. After incubation at 37 C for 24 h, the cells in
according to the manufacturer’s protocol22. Briefly, calibra- the upper chamber were gently wiped off with cotton swabs.
tion of the instrument was performed using 200 nm control The Transwell chamber was fixed with 4% paraformalde-
beads (NanoFCM Inc. China), which were analyzed as a hyde for 30 min, stained with crystal violet dyeing solution
reference for particle concentration. Furthermore, a mixture for 1 min. Images were obtained on an inverted microscope
of beads with different sizes (NanoFCM Inc. China) was (Olympus America, Inc., Center Valley, PA). Each experi-
provided as a reference for size distribution. PBS was used ment was repeated three times, and each time three wells
as a background signal and subtracted from other measured were tested for a sample.
4 Cell Transplantation

Matrigel Tube Formation Assay of HUVECs Statistical Analysis

HUVECs were seeded onto a Matrigel (BD, New Jersey, All data were analyzed using Origin Pro 2020 (OriginLab
USA) precoated 96-well plate at the cell density of Corp., Northampton, MA, USA). The data are presented as
2  104 cells/well, and cultured in ECM for 6 h, in the the means + standard deviation (SD) of at least three inde-
presence of PRP-Exos (50mg /mL) or not. The tube-like pendent simples. Two-tailed Student’s t-test or one-way
structures were pictured using an inverted microscope analysis of variance was used for evaluating the significance
(Olympus America, Inc., Center Valley, PA). The angio- of differences between subgroups. The difference between
genic property was assessed by measuring the total samples was considered significant when P < 0.05.
branching length from three random microscopic fields
using Image J 1.42q software (National Institutes of
Results
Health). Each assay was repeated at least three times. One
representative of three independent experiments was In the present study, an efficient and simple protocol of
shown. activation PRP to release exosomes has been established.
What’s more, exosomes disengaged utilizing calcium gluco-
nate, thrombin, the mixture of calcium gluconate and throm-
bin were successfully compared on morphologic features,
Western Blot Assay particle size distribution, surface charge, and encapsulated
The total protein was extracted using a Total Exosome RNA protein level using electron microscopy, Nanoflow, and WB
& Protein Isolation Kit (Thermo Fisher Scientific, Inc. Wal- analysis, individually. The average platelet count was
tham, MA, USA). The protein concentration was determined (1532 + 81)  109 cell/L in PRP, while the average platelet
using the Pierce™ BCA Protein Assay Kit (Thermo Fisher count was (236 + 22)  109 cell/L in donor venous blood.
Scientific, Waltham, MA, USA). Cell or exosome lysates By using our protocol, those numbers for nucleated cells,
were diluted at a ratio of 1:4 with protein loading buffer such as leukocytes or monocytes, did not surpass one in the
(5) and denatured at 95 C for 5 min. Samples (60 mg of 1  106 platelet tests.
total protein) were separated by SDS-polyacrylamide gels
(5% stacking gel and 12% separation gel) at 120 V for 1 h Different Agonists Influenced PRP Activation
and blotted onto a polyvinylidene difluoride (PVDF) mem-
In the thrombin- and calcium gluconate-activated PRP
brane (Millipore, Jaffrey, NH, USA) for 90 min at 210 mA.
group, the region of fibrin staining by HE in the PRG shaped
The membrane was blocked with 5% weight/volume non-fat
by PRP activation after 90 min has been extensive and
milk powder (BSA, Sangon Biotech, Shanghai, China) in
thickly dispersed. The area of fibrin in the PRG shaped by
TBST (0.1 M Tris-HCl PH 8, 1.5 M NaCl, and 1% Tween-
PRP activation after 90 min in the group of the thrombin
20) for 1 h at RT. The membrane was probed with primary
actuation assembly may have been smaller (P < 0.05), yet
antibodies overnight at 4 C [anti-CD9, anti-CD63, anti-
the area in the PRG shaped by PRP actuation after 90 min in
CD41, anti-CD81, anti-TSG101, anti-flotillin, anti-VEGF,
the group of calcium gluconate activation assembly may
anti-PDGFBB, anti-Calnexin (1:1,000 dilution; Abcam,
have been small, even dispersed. Nevertheless, the control
Cambridge, MA, USA); anti-TGF-b, anti-bFGF, anti-AKT,
group (saline treatment) did not form a fibrin clot (Supple-
anti-p-AKT, anti-p-ERK1/2, anti-ERK1/2, anti-b-actin
mental Fig. S1).
(1:1000 dilution, Cell Signaling Technology, Beverly, MA,
USA)]. After rinsing three times using TBST, we incubated
the immunocomplexes with horseradish peroxidase (HRP)- PRP-Exos Harvested by Gradient Centrifugation
conjugated secondary antibodies (1:4000 dilution, Cell Sig- A protocol for the production of PRP-Exos starting in
naling Technology, Beverly, MA, USA) for 1 h at RT. vein blood is shown in Fig. 1A,B. After separating the
We detected immunoreactivity via the chemiluminescence exosomes, the total protein concentration in all groups
method using a ChemiDoc™ MP Imaging System (Bio-Rad, was measured using the BCA colorimetric protein assay,
Hercules, CA, USA), processed using an enhanced- in which exosomes were lysed with RIPA buffer, and the
chemiluminescence system (ECL, Pierce, Rockford, USA). results are shown in Fig. 1C. The protein concentration of
The densitometric quantification on the protein bands was PRP-Exos obtained in the three stimuli groups was sig-
performed using Image J 1.42q software (National Institutes nificantly higher than that of the control group (saline
of Health). The proteins encapsulated into exosomes were treatment). The protein concentration obtained in the
analyzed by classical western blotting to test for specific calcium-activated group was higher than that in the
exosome markers, such as CD9, CD63, CD81, TSG101, thrombin-activated group (P < 0.01). Although the pro-
flotillin, and the source marker CD41. Calnexin was used tein concentration obtained in the mixture group (calcium
as a negative control. b-actin served as the internal reference. combined with thrombin) was higher than the calcium-
For detailed antibody information, please see Supplementary ionophore group, there was no significant statistical
Table S1. difference (P > 0.05).
Rui et al 5

Figure 1. The workflow of PRP-Exos isolation protocol, starting from PRP and comparison of the total exosome protein concentration.
Exosomes from PRP were isolated using UC. (A) The recommended protocol for PRP preparation and activation. (B) The protocol used in
this study for PRP-Exos extraction. The precipitates in the final step were the obtained PRP-Exos, which were dissolved in sterile PBS.
(C) The total protein levels were estimated in intact exosomes by the BCA protein assay. The highest PRP-Exos yield was detected for the
mixture of calcium gluconate and thrombin, while saline yielded the lowest concentration of exosomes. Statistical significance was deter-
mined using an independent-sample t-test. Data are shown as mean + SD of values of three measurements in each group. n ¼ 3 independent
samples. N.S., not significant. *P < 0.05, and **P < 0.01. UC, ultracentrifugation; PRP, platelet-rich plasma; PRP-Exos, exosomes derived from
platelet-rich plasma; PBS, phosphate-buffered saline.
6 Cell Transplantation

Figure 2. Representative transmission electron microscopy (TEM) images of PRP-Exos. TEM reveals the structure of exosomes. Images
obtained from the exosome population generated by platelets exposed to different agonists: (A) saline; (B) thrombin; (C) calcium gluconate;
and (D) mixture of calcium gluconate and thrombin. The left side of each group are TEM images and the right side is the particle size
distribution diagram of the exosomes, which were counted in the TEM images by Image J. Date presented as mean + SD; n¼3 independent
samples. Scale bar as shown. PRP-Exos, exosomes derived from platelet-rich plasma.

The Difference of PRP-Exos Structure with Different activators varies among preparations. The highest concen-
Agonists tration of particles was observed when PRP-Exos were
obtained by the mixture [(7.16 + 0.46)  1010 particles/
As shown in Fig. 2, the exosomes were isolated from PRP mL; P < 0.05 compared to the calcium-, thrombin-, and
without membrane breakage. The exosomes were ascer- saline-activated groups]. PRP treated with saline yielded the
tained with typical cup-shaped vesicles, measuring lowest concentration of particles [(7.52 + 0.19)  109 par-
30–150 nm in diameter (Fig. 2). However, the size and shape ticles/mL]. The calcium-activated group harvested a higher
of exosomes are different from PRP activated by different concentration of particles [(5.85 + 0.43)  1010 particles/
agonists. Through TEM observations, a few of MVBs were mL] than the saline-activated group (P < 0.01), as did the
present in the calcium-activated group, which contains exo- thrombin-activated group [(4.87 + 0.15)  1010 particles/
somes. The surfaces of PRP-Exos were also rougher in the mL; P < 0.01]. Also, the concentration of particles in the
calcium-activated group than that of the other groups. The calcium-activated group was slightly higher than the
size of PRP-Exos (Fig. 2A, C) was significantly larger than thrombin-activated group (P < 0.05).
the other two groups (Fig. 2B, D). The number of PRP-Exos
activated by saline (Fig. 2A) was the lowest, but the number
of other groups could not be identified under the microscope. The Differences of Exosomal Surface Biomarker and
the Encapsulated Proteins with Different Agonists
The Size Distribution and Concentration of PRP-Exos To further investigate the difference between exosomal sur-
Significantly Vary Among Preparations from Different face proteins and the encapsulated proteins, the purity of
exosomes was examined by immunoblotting for specific
Agonists. marker proteins. CD81, TSG101, CD63, Flotillin, CD9 are
To further determine the size and concentration of PRP-Exos specific markers for exosomes. The source marker (CD41)
obtained by different stimuli, the Nanoflow assay was per- and negative control (calnexin) proteins were also detected
formed for PRP-Exos. As shown in Fig. 3, the average size of by immunoblotting. PRP-Exos activated by different stimuli
exosomes is (78.98 + 4.62) nm. The size of calcium- showed different signal intensities for specific marker pro-
activated PRP-Exos (85.16 + 0.59 nm) was larger than the teins, with the highest intensity detected for prep from the
thrombin-activated group (79.81 + 0.43 nm) and the mixture group (Fig. 4). In contrast, only faint signals for
mixture-activated group (76.00 + 0.37 nm), which showed specific marker and negative control proteins in the saline
broader size distribution with a shift toward the larger size. treatment group were identified, which indicated lower pur-
Furthermore, PRP activated by the mixture released the ity and fewer numbers of exosomes. The signal intensities of
highest concentration of exosomes, followed by the the calcium-activated group were stronger than that of the
calcium-activated group and then the thrombin-activated thrombin-activated group (P < 0.01). The platelet marker
group. The concentration of particles obtained by different CD41 confirmed the exosome identity. Negative control
Rui et al 7

Figure 3. Size and concentration of exosomes from Nanoflow analysis. The size and concentration of the exosome population derived from
platelets exposed to different agonists: (A) saline; (B) thrombin; (C) calcium gluconate; and (D) mixture of calcium gluconate and thrombin.
(E) Concentration (particles/mL) of exosomes was determined by Nanoflow analysis. Statistical significance was determined using an
independent-sample t-test. Data are presented as mean + SD; n ¼ 3 independent samples. N.S., not significant. *P < 0.05, and
**P < 0.01. PRP-Exos, exosomes derived from platelet-rich plasma.

(calnexin), which was found to be present in cell lysates, formation of vessel-like structures. PRP-Exos were interna-
acted as the negative control. These data suggested that the lized by HUVECs, which were mainly located in the cyto-
nanoparticles were exosomes and the WB results were con- plasm of endothelial cells (Fig. 6A). The proliferation of
sistent with the results of the Nanoflow analysis. HUVECs cultured in different concentrations of PRP-Exos
Considering that platelet basic cytokines are mainly pres- and PRP-AS is shown in Fig. 6B, C. The PRP-Exos had a
ent in exosomes16,19. WB analysis was conducted to further concentration-dependent effect on cell proliferation
confirm the differences between cytokine-encapsulated pro- (Fig. 6B), in which HUVECs cultured in 50 mg/mL PRP-
teins in each group. The results are presented in Fig. 5. The Exos activated by saline proliferated significantly better than
results from the WB show that VEGF, PDGFBB, bFGF, and other groups at 24 h (P < 0.05). 50 mg/mL PRP-Exos acti-
TGF-b mainly exist in PRP-Exos, not PRP-AS (P < 0.05). vated by the mixture have a stronger proliferation capacity,
The protein levels of VEGF, PDGFBB, bFGF, TGF-b in the compared to the other groups (Fig. 6C) at all time points
mixture-activated PRP-Exos group were higher than that of tested (P < 0.05). Transwell and tube formation assays
the other groups (P < 0.05). There were no statistical differ- showed that the PRP-Exos activated by the mixture more
ences in the protein content of VEGF, bFGF, and TGF-b significantly promoted migration, formation of vessel-like,
between the calcium- and thrombin-activated groups compared to control, PRP-AS, calcium-activated PRP-Exos,
(P > 0.05). The protein content of PDGFBB in the thrombin-activated PRP-Exos at the same total protein level
thrombin-activated group was higher than that of the (P < 0.05) (Fig. 6D–F). However, there was no statistically
calcium-activated group but lower than that of the mixture significant difference between the calcium-activated PRP-
group (Fig. 5B). Exos group and the thrombin-activated PRP-Exos group
enhancing the formation of vessel-like structures. Figure 5
indicated that VEGF, bFGF, and PDGFBB were enriched in
The Function Assays of PRP-Exos on HUVECs the PRP-Exos, suggesting that AKT and ERK pathways can
Considered that PRP-Exos promote tissue wound repair and be activated by PRP-Exos. To further evaluate the perfor-
endothelial cells play an important role in both angiogenesis mance differences between the different stimulus types, we
and tissue regeneration, the effect of PRP-Exos on HUVECs examined the levels of AKT and ERK phosphorylation by
in vitro was examined. Western Blot following treatment with PRP-Exos or
PRP-Exos were internalized by HUVECs, stimulated PRP-AS for 6 h (Fig. 6G–I). Compared to PRP-AS, incuba-
their proliferation, promoted migration, stimulated in vitro tion with PRP-Exos resulted in a significant increase in
8 Cell Transplantation

Figure 4. Western blot analysis of the surface biomarkers: CD81, TSG101, CD63, Flotillin, CD 9 on PRP-Exos; the source marker CD41;
and the negative control (calnexin), which were found to be present in cell lysates. Densitometric analysis of western blot results: (A)
representative western blot images and (B–E) densitometric analysis. Each lane was loaded with 60 mg of total exosomal protein. Statistical
significance was determined using an independent-sample t-test. Data are shown as mean + SD of values of three measurements in each
group. n ¼ 3 independent samples. N.S., not significant. *P < 0.05, and **P < 0.01 versus PRP-AS; #P < 0.05 and ##P < 0.01.
Rui et al 9

Figure 5. Western blot analysis of PRP-Exos-encapsulated proteins. Representative western blot showing VEGF, PDGFBB, bFGF, TGF-b,
and densitometric analysis of western blot results: (A) representative images; (B–E) densitometric analysis. Each lane was loaded with 60 mg
of total exosomal protein. Statistical significance was determined using an independent-sample t-test. Data are shown as mean + SD of
values of three measurements in each group. n ¼ 3 independent samples. N.S., not significant. *P < 0.05 and **P < 0.01 versus PRP-AS;
#
P < 0.05, and ##P < 0.01.

phosphorylation of AKT and ERK (Fig. 6G). The mixture Different platelet activations may affect PRP-Exos pro-
PRP-Exos group has the highest level of AKT and ERK tein profiles and roles in intercellular communication27–30.
phosphorylation compared with all other groups (Fig. 6G–I). When being used in clinical applications, methods of acti-
vating PRP to release exosomes should be rapid and stan-
dardized, produce a high yield of pure exosomes from small
Discussion PRP volumes, and be compatible with a large number of
PRP has been widely used in the field of repair and regen- samples. A large number of comparative studies have been
eration because it offers many advantages, including easy performed, which compare different activators to activate
applicability, low cost, and an autologous nature 23,24. How- platelets to release exosomes20,27,28,30,31. However, these
ever, the life span of platelets is only around 7–10 days when activation steps are cumbersome. It is necessary to isolate
circulating in the blood and they are not easy to store 25. platelets from PRP and then activate the platelets to release
In general, autologous platelets do not often meet clinical the exosomes, and these activators are not commonly used in
application, and allogeneic platelets have the risk of immune clinical applications. Therefore, it is necessary to compare
rejection. The mechanism by which PRP produces improve- the ability of common drugs (thrombin or calcium gluco-
ments in the field of repair and regeneration remains subject nate) to activate PRP to release exosomes. The methods
to debate5. With the continuous growth of research on cel- studied here are straightforward and easy to perform and sev-
lular exosomes, the potential of exosomes has gained eral samples can be processed simultaneously in short periods.
increasing attention. Exosomes have the following charac- All the activation methods were suitable for PRP-Exos
teristics: low immunogenicity, easy storage, stability, easy isolation, as demonstrated by TEM (cup-shaped vesicles),
mass production, and controllability11,14,26. Exosomes have particle size, and WB detection of the exosome markers:
a high degree of biocompatibility and immune inertness and CD81, TSG101, CD63, Flotillin, and CD915. In particular,
can cross the body’s thick tissue barriers, such as the blood– calcium gluconate alone was found to be weaker than throm-
brain barrier (BBB). Under normal circumstances, exosomes bin or thrombin and calcium gluconate together in PRP acti-
can be directly used as potential therapeutic agents for vation, which was consistent with the previous studies that
immune regulation and tissue repair. investigated the physiological strength of different agonists
10 Cell Transplantation

Figure 6. PRP-Exos effect on HUVECs. (A) representative confocal microphotography of PKH-26 labeled exosomes internalized by
HUVECs; (B-C) CCK8 results of HUVECs treated by PRP-Exos and PRP-AS; (D–F) Transwell migration assay and tube foramtion results
of HUVECs with different treatments. HUVECs were treated with PBS (control), 50 mg/mL PRP-AS, 50 mg/mL PRP-Exosþsaline, 50 mg/mL
PRP-Exosþthrombin, 50 mg/mL PRP-ExosþCa2þ, 50 mg/mL PRP-Exosþmixture; (G–I) Western blot analysis of HUVECs with different
treatments. HUVECs were treated with PBS (control), 50 mg/mL PRP-AS, 50 mg/mL PRP-Exosþsaline, 50 mg/mL PRP-Exosþthrombin,
50 mg/mL PRP-ExosþCa2þ, 50 mg/mL PRP-Exosþmixture. Statistical significance was determined using an independent-sample t-test. Data
are shown as mean + SD of values of three measurements in each group. n ¼ 3 independent samples. N.S., not significant. *P < 0.05, and
**P < 0.01 versus Control; #P < 0.05 and ##P < 0.01.

for platelet activation20,27. TEM showed that some exosomes entry, cytoskeletal remodeling, calpain/caspase activity, and
activated by calcium had rough surfaces, which may be mitochondrial depolarization during activation34. The reason
attributed to the formation of a calcium citrate chelate caus- may be that non-physiological agonists, such as Ca2þ-iono-
ing a partial effect of low pH on the exosome surface pro- phore or detergents, induce most microvesiculation, acting
teins32,33. However, they could be successfully isolated as relaxants27,30.
without membrane breakage, and their morphological fea- The protein content on the exosomal surface was detected
tures were similar in each group. by WB and was consistent with the concentration of exo-
Moreover, the concentration of PRP-Exos activated by somes detected by Nanoflow. Previous studies demonstrated
thrombin alone was found to be lower than that of calcium that biologically-active growth factors are mainly present in
gluconate or thrombin and calcium gluconate together. To PRP-Exos, which have concentration-dependent biological
our knowledge, signaling originates from specific receptors effects16–18. Our study showed that the basic cytokines were
followed by common vesiculation processes, such as Ca2þ almost exclusively present in PRP-Exos and their
Rui et al 11

concentrations varied depending on the PRP activation Authors Contribution

method. PRP-Exos activated by thrombin and calcium glu- SR, YY, WD, CD, and LL conducted and collected the study. SR,
conate together were found to contain more cytokines than WD, YY, HW, YF and PS analyzed the data and contributed to the
groups activated by thrombin or calcium gluconate alone. discussion. SR, DGA, and WD wrote the manuscript, contributed to
To further understand functional differences in angiogen- the discussion, reviewed, and edited the manuscript. All authors
esis between all groups, the effect of PRP-Exos on HUVECs read and approved the final manuscript. WD and LL took full
was examined, which play an important role in both angio- responsibility for this study.
genesis and tissue regeneration35. Results showed that 50 mg/
Ethical Approval
mL PRP-Exos activated by thrombin and calcium gluconate
together could significantly promote HUVECs proliferation, The ethical approval to report this work was obtained from the
increase HUVECs migration, promote the formation of Ethical Committee Board of Southwest Hospital.
vessel-like by HUVECs via the AKT ERK signal pathway,
Statement of Human and Animal Rights
compared with those in the other groups. While the down-
All procedures in this study were conducted following the protocol
stream signaling kinases AKT and ERK were activated by
approved by “the Ethical Committee Board of Southwest Hospital”
PRP-Exos, it appeared that AKT was a major one since much
and complied with the recommendations of the Declaration of
higher levels of p-AKT than that of p-ERK were observed. Helsinki.
Although PRP-Exos activated by thrombin or thrombin and
calcium gluconate together contained more PDGFBB than Statement of Informed Consent
the other groups, PRP-Exos activated by thrombin did not Written informed consent was obtained from the patients for their
promote HUVECs proliferation, increase HUVECs migration, anonymized information to be published in this article.
promote the formation of vessel-like by HUVECs, enhance
activation of AKT and ERK signal pathway more dramatically Declaration of Conflicting Interests
than PRP-Exos activated by calcium gluconate. It is suggested The author(s) declared no potential conflicts of interest with respect
that calcium might perform a cell-intrinsic role in regulating to the research, authorship, and/or publication of this article
the expression of certain important molecules by activating
PRP. Relevant studies showed that the lipid growth factors of Funding
platelet microparticles may be the main active factor of pro- The author(s) disclosed receipt of the following financial support
moting angiogenesis, while the protein component may be the for the research, authorship, and/or publication of this article: This
secondary factor36,37. For instance, Sphingosine 1-phosphate research was funded by Natural Science Foundation of Chongqing
(S1P), one of these platelet-derived lipid growth factors, has Municipal Science and Technology Bureau (cstc2018jcyjAX0335;
been demonstrated that mediates cytoprotection, proliferation, cstc2020jcyj-msxmX0298); the Fundamental Research Funds of
migration, and tube formation via activating PI3K/AKT38,39. Central Universities at Chongqing University (2019CDYGYB020)
awarded to Dr. Wuquan Deng. This work was funded by the Nat-
In summary, we compared the effects of different ago-
ural Science Foundation of Science, and Basic Research and Fron-
nists on the quality and quantity of PRP-Exos and the effect
tier Exploration of Science and Technology Commission by
of PRP-Exos on HUVECs. We found that all agonists could Chongqing Municipality (Grant No.cstc2018jcyjAX0788) awarded
activate platelets to release exosomes, but the activation to Dr.Yan Chen. This study is partially supported by National
efficiency and cytokine content are different. Calcium glu- Institutes of Health, National Institute of Diabetes and Digestive
conate activation can release more exosomes, and thrombin and Kidney Diseases Award Number 1R01124789-01A1 awarded
activation can obtain exosomes containing high concentra- to Prof. DG Armstrong.
tions of PBDGBB. The effects from activating PRP with
thrombin and calcium gluconate together are better than ORCID iD
using thrombin or calcium alone, both in quality and quantity Wuquan Deng https://orcid.org/0000-0002-5450-9713
of exosomes. 50 mg/mL PRP-Exos activated by thrombin
and calcium gluconate together could more significantly Supplemental Material
promote HUVECs proliferation, increase HUVECs migra- Supplemental material for this article is available online.
tion, promote the formation of vessel-like by HUVECs via
the AKT ERK signal pathway, compared with other groups. References
Our studies showed that the number of PRP-Exos and the 1. Lo Monaco M, Gervois P, Beaumont J, Clegg P, Bronckaers A,
content of cytokines in the PRP-Exos, PRP-Exos biological Vandeweerd J-M, Lambrichts I. Therapeutic potential of dental
effects on HUVECs are different when PRP is activated by pulp stem cells and leukocyte- and platelet-rich fibrin for
different agonists. This study provides a protocol for selecting osteoarthritis. Cells. 2020;9(4):980.
the appropriate PRP activator to obtain high-quality exosomes 2. Andriolo L, Altamura SA, Reale D, Candrian C, Zaffagnini S,
for future applications. However, the mechanism behind the Filardo G.Nonsurgical treatments of patellar tendinopathy:
difference in the number of exosomes and cytokines the effect multiple injections of platelet-rich plasma are a suitable option:
of PRP-Exos on HUVECs, generated by different activators, a systematic review and meta-analysis. Am J Sports Med.
are not clear, and further exploration should be performed. 2019;47(4):1001–1018.
12 Cell Transplantation

3. Laudy AB, Bakker EW, Rekers M, Moen MH. Efficacy of of chronic cutaneous wounds via activation of YAP in a dia-
platelet-rich plasma injections in osteoarthritis of the knee: betic rat model. Theranostics. 2017;7(1):81–96.
a systematic review and meta-analysis. Br J Sports Med. 18. Tao SC, Yuan T, Rui BY, Zhu ZZ, Guo SC, Zhang CQ. Exosomes
2015;49(10):657–672. derived from human platelet-rich plasma prevent apoptosis
4. Martinez-Zapata MJ, Martı́-Carvajal AJ, Solà I, Expósito JA, induced by glucocorticoid-associated endoplasmic reticulum
Bolı́bar I, Rodrı́guez L, Garcia J, Zaror C. Autologous platelet- stress in rat osteonecrosis of the femoral head via the Akt/Bad/
rich plasma for treating chronic wounds. Cochrane Database Bcl-2 signal pathway. Theranostics. 2017;7(3):733–750.
Syst Rev. 2016;(5):Cd006899. 19. Brill A, Dashevsky O, Rivo J, Gozal Y, Varon D. Platelet-derived
5. Gentile P, Calabrese C, De Angelis B, Dionisi L, Pizzicannella microparticles induce angiogenesis and stimulate post-ischemic
J, Kothari A, De Fazio D, Garcovich S. Impact of the different revascularization. Cardiovasc Res. 2005;67(1):30–38.
preparation methods to obtain autologous non-activated 20. Aatonen MT, Ohman T, Nyman TA, Laitinen S, Grönholm M,
platelet-rich plasma (A-PRP) and activated platelet-rich Siljander PR. Isolation and characterization of platelet-derived
plasma (AA-PRP) in plastic surgery: wound healing and hair extracellular vesicles. J Extracell Vesicles. 2014;3.
regrowth evaluation. Int J Mol Sci. 2020;21(2):431. 21. Théry C, Amigorena S, Raposo G, Clayton A. Isolation and
6. Deng W, Boey J, Chen B, Byun S, Lew E, Liang Z, Armstrong characterization of exosomes from cell culture supernatants
DG. Platelet-rich plasma, bilayered acellular matrix grafting and biological fluids. Curr Protoc Cell Biol. 2006;Chapter 3:
and negative pressure wound therapy in diabetic foot infection. Unit 3.22.
J Wound Care. 2016;25(7):393–397. 22. Tian Y, Ma L, Gong M, Su G, Zhu S, Zhang W, Wang S, Li Z,
7. Li T, Ma Y, Wang M, Wang T, Wei J, Ren R, He M, Wang G, Chen C, Li L, Wu L, et al. Protein profiling and sizing of
Boey J, Armstrong DG, Deng W, et al. Platelet-rich plasma extracellular vesicles from colorectal cancer patients via flow
plays an antibacterial, anti-inflammatory and cell proliferation- cytometry. ACS Nano. 2018;12(1):671–680.
promoting role in an in vitro model for diabetic infected 23. Sun Y, Feng Y, Zhang CQ, Chen SB, Cheng XG. The regen-
wounds. Infect Drug Resist. 2019;12:297–309. erative effect of platelet-rich plasma on healing in large osteo-
8. Jiang X, Li N, Yuan Y, Yang C, Chen Y, Ma Y, Wang J, Du D, chondral defects. Int Orthop. 2010;34(4):589–597.
Boey J, Armstrong DG, Deng W. Limb salvage and prevention 24. Foster TE, Puskas BL, Mandelbaum BR, Gerhardt MB, Rodeo
of ulcer recurrence in a chronic refractory diabetic foot osteo- SA. Platelet-rich plasma: from basic science to clinical appli-
myelitis. Diabetes Metab Syndr Obes. 2020;13:2289–2296. cations. Am J Sports Med. 2009;37(11):2259–2272.
9. He M, Guo X, Li T, Jiang X, Chen Y, Yuan Y, Chen B, Yang 25. van der Meijden PEJ, Heemskerk JWM. Platelet biology and
G, Fan Y, Liang Z, Armstrong DG, et al. Comparison of allo- functions: new concepts and clinical perspectives. Nat Rev
geneic platelet-rich plasma with autologous platelet-rich Cardiol. 2019;16(3):166–179.
plasma for the treatment of diabetic lower extremity ulcers. 26. ELA S, Mager I, Breakefield XO, Wood MJ. Extracellular
Cell Transplant. 2020;29:963689720931428. vesicles: biology and emerging therapeutic opportunities. Nat
10. Shao S, Pan R, Chen Y. Autologous platelet-rich plasma for Rev Drug Discov. 2013;12(5):347–357.
diabetic foot ulcer. Trends Endocrinol Metab. 2020;31(12): 27. Aatonen M, Grönholm M, Siljander PR. Platelet-derived
885–890. microvesicles: multitalented participants in intercellular
11. Théry C, Zitvogel L, Amigorena S. Exosomes: composition, communication. Semin Thromb Hemost. 2012;38(1):
biogenesis and function. Nat Rev Immunol. 2002;2(8): 102–113.
569–579. 28. Perez-Pujol S, Marker PH, Key NS. Platelet microparticles are
12. Heijnen HF, Schiel AE, Fijnheer R, Geuze HJ, Sixma JJ. Acti- heterogeneous and highly dependent on the activation mechan-
vated platelets release two types of membrane vesicles: micro- ism: studies using a new digital flow cytometer. Cytometry A.
vesicles by surface shedding and exosomes derived from 2007;71(1):38–45.
exocytosis of multivesicular bodies and alpha-granules. Blood. 29. Shai E, Rosa I, Parguiña AF, Motahedeh S, Varon D, Garcı́a Á.
1999;94(11):3791–3799. Comparative analysis of platelet-derived microparticles
13. van Niel G, D’Angelo G, Raposo G. Shedding light on the cell reveals differences in their amount and proteome depending
biology of extracellular vesicles. Nat Rev Mol Cell Biol. 2018; on the platelet stimulus. J Proteomics. 2012;76:287–296.
19(4):213–228. 30. Milioli M, Ibáñez-Vea M, Sidoli S, Palmisano G, Careri M,
14. Raposo G, Stoorvogel W. Extracellular vesicles: exosomes, Larsen MR. Quantitative proteomics analysis of platelet-
microvesicles, and friends. J Cell Biol. 2013;200(4):373–383. derived microparticles reveals distinct protein signatures when
15. Pegtel DM, Gould SJ. Exosomes. Annu Rev Biochem. 2019; stimulated by different physiological agonists. J Proteomics.
88:487–514. 2015;121:56–66.
16. Torreggiani E, Perut F, Roncuzzi L, Zini N, Baglı̀o SR, Baldini 31. Guo J, Feng C, Zhang B, Zhang S, Shen X, Zhu J, Zhao XX.
N. Exosomes: novel effectors of human platelet lysate activity. Extraction and identification of platelet-derived microparti-
Eur Cell Mater. 2014;28:137–151; discussion 151. cles. Mol Med Rep. 2019;20(3):2916–2921.
17. Guo SC, Tao SC, Yin WJ, Qi X, Yuan T, Zhang CQ. Exosomes 32. Yamauchi M, Shimizu K, Rahman M, Ishikawa H, Takase H,
derived from platelet-rich plasma promote the re-epithelization Ugawa S, Okada A, Inoshima Y. Efficient method for isolation
Rui et al 13

of exosomes from raw bovine milk. Drug Dev Ind Pharm. 37. McVey MJ, Weidenfeld S, Maishan M, Spring C, Kim M,
2019;45(3):359–364. Tabuchi A, Srbely V, Takabe-French A, Simmons S, Arenz
33. Ban JJ, Lee M, Im W, Kim M. Low pH increases the yield of C, Semple JW, et al. Platelet extracellular vesicles mediate
exosome isolation. Biochem Biophys Res Commun. 2015; transfusion-related acute lung injury by imbalancing the sphin-
461(1):76–79. golipid rheostat. Blood. 2021;137(5):690–701.
34. Morel O, Jesel L, Freyssinet JM, Toti F. Cellular mechanisms 38. Tawa H, Rikitake Y, Takahashi M, Amano H, Miyata M,
underlying the formation of circulating microparticles. Arter- Satomi-Kobayashi S, Kinugasa M, Nagamatsu Y, Majima T,
ioscler Thromb Vasc Biol. 2011;31(1):15–26. Ogita H, Miyoshi J, et al. Role of afadin in vascular endothelial
35. Wang C, Wang M, Xu T, Zhang X, Lin C, Gao W, Xu H, Lei B, growth factor- and sphingosine 1-phosphate-induced angio-
Mao C. Engineering bioactive self-healing antibacterial exo- genesis. Circ Res. 2010;106(11):1731–1742.
somes hydrogel for promoting chronic diabetic wound healing 39. Heller R, Chang Q, Ehrlich G, Hsieh SN, Schoenwaelder SM,
and complete skin regeneration. Theranostics. 2019;9(1):65–76. Kuhlencordt PJ, Preissner KT, Hirsch E, Wetzker R. Overlap-
36. Kim HK, Song KS, Chung JH, Lee KR, Lee SN. Platelet ping and distinct roles for PI3Kbeta and gamma isoforms in
microparticles induce angiogenesis in vitro. Br J Haematol. S1P-induced migration of human and mouse endothelial cells.
2004;124(3):376–384. Cardiovasc Res. 2008;80(1):96–105.