Characterization of inflammatory factor-induced changes in mesenchymal stem cell exosomes and sequencing analysis of exosomal microRNAs.

BACKGROUND: Treatments utilizing stems cells often require stem cells to be exposed to inflammatory environments, but the effects of such environments are unknown. AIM: To examine the effects of inflammatory cytokines on the morphology and quantity of mesenchymal stem cell exosomes (MSCs-exo) as well as the differential expression of microRNAs (miRNAs) in the exosomes.
METHODS: MSCs were isolated from human umbilical tissue by enzymatic digestion. Exosomes were then collected after a 48-h incubation period in a serum-free medium with one of the following the inflammatory cytokines: None (control), vascular cell adhesion molecule-1 (VCAM-1), tumor necrosis factor (TNF) α, and interleukin (IL) 6. The morphology and quantity of each group of MSC exosomes were observed and measured. The miRNAs in MSCs-exo were sequenced. We compared the sequenced data with the miRBase and other non-coding databases in order to detect differentially expressed miRNAs and explore their target genes and regulatory mechanisms. In vitro tube formation assays and Western blot were performed in endothelial cells which were used to assess the angiogenic potential of MSCs-exo after inflammatory cytokine stimulation.
RESULTS: MSCs-exo were numerous, small, and regularly shaped in the VCAM-1 group. TNFα stimulated MSCs to secrete larger and irregular exosomes. IL6 led to a reduced quantity of MSCs-exo. Compared to the control group, the TNFα and IL6 groups had more downregulated differentially expressed miRNAs, particularly angiogenesis-related miRNAs. The angiogenic potential of MSCs-exo declined after IL6 stimulation.
CONCLUSION: TNFα and IL6 may influence the expression of miRNAs that down-regulate the PI3K-AKT, MAPK, and VEGF signaling pathways; particularly, IL6 significantly down-regulates the PI3K-AKT signaling pathway. Overall, inflammatory cytokines may lead to changes in exosomal miRNAs that abnormally impact cellular components, molecular function, and biological processes. ©The Author(s) 2019. Published by Baishideng Publishing Group Inc. All rights reserved.

Core tip: The morphology and quantity of mesenchymal stem cell exosomes (MSCs-exo) are impacted in different inflammatory cytokine environments. Inflammatory cytokines impair the ability of MSCs-exo to promote angiogenesis. For instance, the tumor necrosis factor α and interleukin 6 groups exhibited decreased numbers of angiogenesis-related microRNAs (miRNAs), such as miR-196a-5p, miR-17-5p, miR-146b-5p, miR-21-3p, and miR-320. The same groups also had downregulated angiogenesis-related signaling pathways, such as PI3K-AKT and VEGF. Inflammatory cytokines may lead to changes in exosomal miRNAs that abnormally impact cellular components, molecular function, and biological processes, particularly angiogenesis-related miRNAs.

INTRODUCTION

Stem cell transplantation has been developing rapidly and has resulted in breakthroughs for the treatment of various diseases. There is especially an interest in the transplantation of mesenchymal stem cells (MSCs), which are tissue-derived cells with self-renewal abilities. Their exosomes (MSCs-exo) not only contain the unique active components of all stem cells, but also are relatively more safe, are more chemically stable, and have the capacity for targeted delivery to biological pathways of interest[1-4].

Exosomes were first found in vitro in cultured sheep erythrocyte supernatant[5]. They are membranous vesicles with a diameter of 30-150 nm and a density of 1.10-1.18 g/mL. They are able to affect gene regulation by carrying and releasing various bioactive molecules such as microRNAs (miRNAs) and proteins, both of which can then function as paracrine signaling mediators impacting biological pathways relevant to disease processes[6]. MiRNAs are non-coding RNAs 22-25 nucleotides in length[7-10]. Exosomes are especially important in producing miRNAs that impact angiogenesis[11].

There is promising research on using MSCs-exo to encourage wound healing and to treat inflammatory arthritis and ischemic diseases. However, in treating these diseases, MSCs are exposed to microenvironments filled with numerous inflammatory cytokines, such as vascular cell adhesion molecule-1 (VCAM-1), tumor necrosis factor (TNF) α, and interleukin (IL) 6. The effects of these inflammatory cytokines on the morphology and quantity of MSCs-exo and how these effects impact the production of miRNAs and downstream regulatory mechanisms are largely unknown. In this study, we analyzed the effects of VCAM-1, TNFα, and IL6 on the morphology and quantity of MSCs-exo, how these effects enhance differential expression of miRNAs, and how the target genes of these miRNAs and their associated regulatory mechanisms are regulated.

MATERIALS AND METHODS

Cell culture

Human umbilical mesenchymal stem cells were obtained from the Polywin Corporation (Guangzhou, China). The phenotypes of the MSCs were characterized by flow cytometric analysis of cell surface antigens, including tests for the cluster of differentiation (CD)29, CD34, CD44, CD73, and CD105. The MSCs were divided into four groups: Control group: MSCs (1.0 × 105 cells/mL) were cultured in a serum-free DMEM/F12 (Sigma-Aldrich, United States) medium for 48 h; VCAM-1 group: MSCs (1.0 × 105 cells/mL) were cultured in a serum-free DMEM/F12 (Sigma-Aldrich) medium, and VCAM-1 reagent (ADP5, R&D Systems, United States) was added to the medium at a concentration of 20 ng/mL for 48 h; TNFα group: MSCs (1.0 × 105 cells/mL) were cultured in a serum-free DMEM/F12 (Sigma-Aldrich) medium, and TNFα reagent (T6674, Sigma-Aldrich) was added to the medium at a concentration of 20 ng/mL for 48 h; and IL6 group: MSCs (1.0 × 105 cells/mL) were cultured in a serum-free DMEM/F12 (Sigma-Aldrich) medium, and IL6 reagent (200-06-20, PeproTech, United States) was added to the medium at a concentration 20 ng/mL for 48 h. The number, distribution, and morphology of cells in each group were observed under a microscope at 100× or 200× magnification for 48 h.

Exosome isolation

Exosomes were isolated from the culture supernatant by ultracentrifugation according to methods described previously[12,13]. Briefly, the culture medium of each group was collected and centrifuged at 2000 × g for 10 min at 4 °C. The supernatant was then centrifuged at 10000 × g for 10 min at 4 °C. Next, the supernatant was passed through a 0.2-μm filter (Steradisc; Kurabo, Bio-Medical Department, Tokyo, Japan). The filtrate was ultracentrifuged at 100000 × g for 70 min at 4 °C (Type 70Ti ultracentrifuge; Beckman Coulter, Inc., Brea, CA, United States). The precipitate was next rinsed with phosphate buffered saline (PBS) and ultracentrifuged at 100000 × g for 70 min at 4 °C. The exosome-enriched fraction was next reconstituted in PBS for further studies.

Characteristics and distributions of exosomes in each group were observed. The particle size and concentration of exosomes in each group were measured by nanoparticle tracking analysis.

Differential miRNA analysis and target gene and regulatory signal pathway prediction

The miRNAs of MSCs-exo were sequenced by BGISEQ-500 technology in each group. Sequenced data were compared with miRBase and other non-coding databases. Bioinformatics analysis pipeline steps for miRNA sequencing were: (1) Filtering small RNAs: 18-30 nt RNA segments were separated by polyacrylamide gel electrophoresis (PAGE); (2) 3’ adaptor ligation: A 5-adenylated, 3-blocked single-stranded DNA adapter was linked to the 3’ end of selected small RNAs from step 1; (3) Reverse primer annealing: the RT primer was added to the solution from step 2 and cross-linked to the 3’ adapter of the RNAs and to excess free 3’ adapter; (4) 5’ adaptor ligation: a 5’ adaptor was linked to the 5' end of the product from step 3. The adaptor was attached to the end only, not to the 3’ adaptor or RT primer hybrid chain, thus greatly reducing self-ligation; (5) cDNA synthesis: The RT primers in step 3 were reverse extended to synthesize cDNA strands; (6) PCR amplification: High-fidelity polymerase was used to amplify cDNA, and cDNA with both 3’ and 5’ adaptors was enriched; (7) Library fragment selection: The PCR products of 100-120 bp were separated by PAGE to eliminate primer dimers and other byproducts; (8) Library quantitative and pooling cyclization; (9) Eliminating the low-quality reads, adaptors and other contaminants to obtain clean reads; (10) Summarizing the length distribution of the clean tags, common, and specific sequences between samples; (11) Assigning the clean tags to different categories; (12) Predicting novel miRNAs; (13) Function annotation of known miRNAs; and (14) Comparing clean reads to the reference base group and other small RNA databases using AASRA software [14], except that Rfam was compared with cmsearch[15]. We used TPM[16] to standardize miRNA expression levels and predicted target genes using RNAhybrid[17], miRanda[18], and TargetScan[19].

Hierarchical clustering analysis showed differentially expressed miRNAs by functional pheatmap. The P-values obtained from the differential gene expression tests were corrected by controlling the false discovery rate (FDR)[20] as more stringent criteria with smaller FDRs and bigger fold-change values can be used to identify differentially expressed miRNAs.

Gene ontology (GO) enrichment analysis was performed to identify all GO terms that are significantly enriched in a list of target genes of differentially expressed miRNAs, as well as the genes that correspond to specific biological functions.

The hypergeometric test was then used to find significantly enriched GO terms based on this database (http://www.geneontology.org/). Pathway-based analyses were used to discover the biological functions of target genes using KEGG[21] (the major public pathway-related database).

In vitro Matrigel tube formation assay

HUVECs (4.0 × 104, serum-starved overnight) were seeded in a 96-well plate, cultured in 5% CO2 overnight, and then treated with PBS (control), control scrambled MSCs-exo, MSCs-exoTNFα (5 × 109, stimulated with TNFα), or MSCs-exoIL6 (5 × 109, stimulated with IL6). The plates were previously coated with 150 μL of growth factor-reduced Matrigel (356234, Corning, United States) in serum-free medium. Tube formation ability of control or MSCs-exo-treated HUVECs was examined by determining the total number of tubes formed and branching points in 4 to 6 h. Each condition in each experiment was assessed at least in duplicate.

Western blot analysis

HUVECs (4.0 × 104) were treated with PBS (control), control scrambled MSCs-exo, MSCs-exoTNFα (5 × 109, stimulated with TNFα), or MSCs-exoIL6 (5 × 109, stimulated with IL6). The effect of MSCs-exo treatment on PI3K-AKT and MAPK, which are related to angiogenic signaling, was examined by measuring the expression of AKT (1:1000 dilution, 4691S, Cell Signaling Technology, Danvers, MA, United States), phospho-AKT (1:1000 dilution, 13038S, Cell Signaling Technology), phospho-p44/42 MAPK (Erk1/2) (1:2000 dilution, 4370S, Cell Signaling Technology), and p44/42 MAPK (Erk1/2) (1:1000 dilution, 4695S, Cell Signaling Technology) in endothelial cells by Western blot. Each condition in each experiment was assessed at least in triplicate.

Statistical analysis

The data from each group were collected and analyzed using SPSS 11.5 software (IBM SPSS China, Shanghai, china). Numerical data are presented as the mean ± SE; comparisons between groups were evaluated by Student’s t-test or ANOVA, with P < 0.05 considered significant.

RESULTS

Phenotypic characterization of MSCs

Cell purity (85% to 95%) was determined via flow cytometry. The cells were positive for mesenchymal cell markers such as CD29, CD44, CD73, and CD105 and negative for hematopoietic cell markers such as CD34 and HLA-DR (Figure 1).

Figure 1

Flow cytometric analysis of mesenchymal stem cell-related cell surface markers. High expression of positive mesenchymal cell markers (CD29, CD44, CD73, and CD105), and low expression of negative cell markers such as CD34 and HLA-DR were observed. n = 3 to 5; P < 0.05,

Effect of inflammatory cytokines on MSCs-exo

The VCAM-1 group had small regularly shaped MSCs-exo. The TNFα group had large irregularly shaped MSCs-exo. The IL6 group had medium regularly shaped MSCs-exo (Figure 2)

Figure 2

Morphology of mesenchymal stem cell exosomes (magnification, ×65000). A-D: Mesenchymal stem cell exosomes (MSCs-exo) in (A) the control group (yellow arrow), (B) the vascular cell adhesion molecule-1 group (yellow arrow), (C) the tumor necrosis factor α group (yellow arrow), and (D) the interleukin 6 group (yellow arrow).

The density of MSCs-exo was 7.42 × 108/mL in the control group, 1.10 × 109/mL in the VCAM-1 group, 7.37 × 108/mL in the TNFα group, and 3.01 × 108/mL in the IL6 group (Figure 3).

Figure 3

Nanoparticle tracking analysis. A-D: Density of exosomes of different sizes in (A) the control group, (B) the vascular cell adhesion molecule-1 group, (C) the tumor necrosis factor α group, and (D) the interleukin 6 group.

Effect of differential miRNA expression, secondary to inflammatory cytokine exposure, on biological function

Correlation analyses showed that the miRNA expression profiles in the IL6 group (0.583) and TNFα group (0.697) were more different from that of the control group than that in the VCAM1 group (0.985) (Figures 4 and 5). The top 10 miRNAs in each group are shown in Table 1. Many miRNAs, particularly some important angiogenesis-related miRNAs, were downregulated in the TNFα group and IL6 group compared to the control group (Tables 2-4 and Figure 6). Hierarchical clustering indicated that the expression levels of the majority of miRNAs in the IL6 group were downregulated compared with those of the control group (Figure 7). According to GO enrichment analysis, miRNAs in exosomes exposed to inflammatory cytokines, compared to controls, had a different regulatory effect on cellular components, molecular function, and biological processes (Figure 8). More specifically, pathway enrichment analysis showed that the target genes of the differentially expressed miRNAs, including those related to angiogenesis, differed among the four groups (Figures 9 and 10). The following angiogenesis-related pathways were more down-regulated in the TNFα and IL6 groups than in the control group: The PI3K-AKT signaling pathway (Q = 0.0978197212 and 0.0581120875 in the TNFα group and IL6 group, respectively), the MAPK signaling pathway (Q = 0.5775485 and 0.9837761532 in the TNFα group and IL6 group, respectively), and the VEGF signaling pathway (Q = 0.4082212190 and 0.1711566 in the TNFα group and IL6 group, respectively) (Figures 10-13).

Figure 4

Correlation analyses between groups. Blue color represents the correlation coefficient (the deeper the blue, the stronger the correlation). VCAM-1: Vascular cell adhesion molecule-1; TNF: Tumor necrosis factor; IL6: Interleukin 6.

Figure 5

Distribution of differentially expressed miRNAs in different groups. VCAM-1: Vascular cell adhesion molecule-1; TNF: Tumor necrosis factor; IL6: Interleukin 6.

Table 1

Differentially expressed miRNAs

Top 10 differentially expressed miRNAs
Control group			VCAM-1 group			In the TNFα group			In the IL6 group
sRNA id	Count (16379497)	TPM	sRNA id	Count (23095609)	TPM	sRNA id	Count (20739574)	TPM	sRNA id	Count (32116232)
miR-21-5p	1544110	76933.99	miR-21-5p	1640334	70885.38	miR-21-5p	1454726	70066.87	miR-21-5p	1323387
miR-127-3p	159283	9724.54	miR-127-3p	259830	11250.19	miR-100-5p	408192	19681.79	miR-199a-3p,miR-199b-3p	317377
miR-199a-3p,miR-199b-3p	139220	8499.65	miR-199a-3p, miR-199b-3p	155376	6727.51	miR-127-3p	321306	15492.41	miR-127-3p	300549
miR-100-5p	132865	8111.67	miR-100-5p	140774	6095.27	miR-199a-3p, miR-199b-3p	271902	13110.3	miR-100-5p	265054
miR-222-3p	89564	5468.06	miR-26a-5p	107800	4667.55	miR-23a-3p	140623	6780.42	miR-34a-5p	178542
miR-26a-5p	77524	4732.99	miR-34a-5p	86786	3757.68	miR-222-3p	129932	6264.93	miR-23a-3p	144385
miR-34a-5p	76681	4681.52	miR-451a	77297	3346.83	miR-146a-5p	128078	6175.54	miR-222-3p	111957
miR-181a-5p	67287	4108	miR-23a-3p	71930	3114.44	miR-34a-5p	119952	5783.73	miR-146a-5p	100038
miR-411-5p	56606	3455.91	miR-196a-5p	63939	2768.45	miR-143-3p	101398	4889.11	miR-26a-5p	98114
miR-143-3p	49739	3036.66	miR-181a-5p	62752	2717.05	miR-26a-5p	99628	4803.76	miR-181a-5p	97685

Count is the number of mapped tags, and the value in parentheses is the total number. Transcripts per kilobase million is the standardized expression value. VCAM-1: Vascular cell adhesion molecule-1; TNF: Tumor necrosis factor; IL6: Interleukin 6; TPM: Transcripts per kilobase million.

Table 2

Top 10 differentially expressed miRNAs in each group (Poisson distribution screening)

Comparison of top 10 differentially expressed miRNAs
VCAM-1 group vs control group									TNFα group vs control group									IL6 group vs control group
miRNA id	Count (control)	Count (VCAM-1)	TPM (control)	TPM (VCAM-1)	Log2 ratio (VCAM-1/control)	Up/down-regulation (VCAM-1/control)	P	FDR	miRNA id	Count (control)	Count (TNFα)	TPM (control)	TPM (TNFα)	Log2 ratio (TNFα/control)	Up/down-regulation (TNFα/control)	P	FDR	miRNA id	Count (control)	Count (IL6)	TPM (control)	TPM (IL6)	Log2 ratio (IL6/control)	Up/down-Regulation (IL6/control)	P value	FDR
miR-146a-5p	20510	5173	1252.18	223.98	-2.483	Down	0	0	miR-146a-5p	20510	128078	1252.18	6175.54	2.302123	Up	0	0	miR-3529-3p	473	8930	28.88	278.05	3.267202	Up	0	0
miR-149-5p	4558	2391	278.27	103.53	-1.42644	Down	0	0	miR-342-3p	2117	7130	129.25	343.79	1.411363	Up	0	0	miR-320c	1276	7983	77.9	248.57	1.673957	Up	0	0
miR-222-3p	89564	48444	5468.06	2097.54	-1.38233	Down	0	0	miR-23a-3p	42777	140623	2611.62	6780.42	1.37643	Up	0	0	miR-24-3p	11151	50968	680.79	1586.99	1.221011	Up	0	0
miR-29a-3p	8451	5054	515.95	218.83	-1.23742	Down	0	0	miR-100-5p	132865	408192	8111.67	19681.79	1.278791	Up	0	0	miR-7-5p	16653	101	1016.7	3.14	-8.33891	Down	0	0
miR-451a	2760	77297	168.5	3346.83	4.311975	Up	0	0	miR-337-3p	5655	16027	345.25	772.77	1.162398	Up	0	0	miR-196a-5p	26485	11440	1616.96	356.21	-2.18248	Down	0	0
miR-490-5p	1839	475	112.27	20.57	-2.44836	Down	2.00E-305	1.20E-303	miR-409-3p	4207	11169	256.85	538.54	1.068127	Up	0	0	miR-451a	2760	1658	168.5	51.62	-1.70675	Down	0	0
miR-1246	1812	6191	110.63	268.06	1.276813	Up	1.39E-276	8.02E-275	miR-221-3p	4067	10541	248.3	508.26	1.033482	Up	0	0	miR-379-5p	13430	8083	819.93	251.68	-1.70391	Down	0	0
miR-181a-2-3p	3400	2008	207.58	86.94	-1.25558	Down	2.31E-219	1.18E-217	miR-7-5p	16653	136	1016.7	6.56	-7.27598	Down	0	0	miR-196b-5p	18735	15676	1143.81	488.1	-1.2286	Down	0	0
miR-204-5p	3116	1757	190.24	76.08	-1.32223	Down	3.36E-218	1.66E-216	miR-196a-5p	26485	8359	1616.96	403.05	-2.00425	Down	0	0	miR-424-5p	12774	11917	779.88	371.06	-1.0716	Down	0	0
miR-378e	2	817	0.12	35.37	8.203348	Up	4.36E-186	2.00E-184	miR-26b-5p	6496	3410	396.59	164.42	-1.27026	Down	0	0	miR-26b-5p	6496	6086	396.59	189.5	-1.06545	Down	0	0

P-values were corrected. TPM: Transcripts per kilobase million; FDR: False discovery rate; VCAM-1: Vascular cell adhesion molecule-1; TNF: Tumor necrosis factor; IL6: Interleukin 6.

Table 4

Comparison of differentially expressed angiogenesis-related miRNAs between the IL6 group and control group

miRNA id	Count (control)	Count (IL6)	TPM (control)	TPM (IL6)	Log2 ratio (IL6/control)	Up/down-regulation (IL6/control)	P	FDR
hsa-miR-196a-5p	26485	11440	1616.96	356.21	-2.182484067	Down	0	0
hsa-miR-17-5p	252	138	15.39	4.3	-1.839584667	Down	2.24E-35	2.98E-34
hsa-miR-146b-5p	78	44	4.76	1.37	-1.79678568	Down	1.08E-11	7.53E-11
hsa-miR-21-3p	440	358	26.86	11.15	-1.268415595	Down	4.66E-35	6.02E-34
hsa-miR-320e	161	153	9.83	4.76	-1.046229843	Down	1.78E-10	1.11E-09

P-values were corrected. TPM: Transcripts per kilobase million; FDR: False discovery rate; IL6: Interleukin 6.

Figure 6

Volcano plots. A-C: Volcano plots of (A) vascular cell adhesion molecule-1 group vs control group, (B) tumor necrosis factor α group vs control group, and (C) interleukin 6 group vs control group. The X axis is the log2(Fold change), and the Y axis is the -log10(FDR), with green points indicating down-regulation [log2(Fold old change) ≤ -1 and FDR ≤ 0.001] and red points indicating upregulation [log2(Fold change) ≥ 1 and FDR ≤ 0.001]. FDR: False discovery rate.

Figure 7

Hierarchical clustering of differentially expressed miRNAs. A: Vascular cell adhesion molecule-1 group vs control group; B: Tumor necrosis factor α group vs control group; C: Interleukin 6 group vs control group. The X axis represents each pair of differences, and the Y axis represents differentially expressed miRNAs. The colors indicate the fold change, with red showing up-regulation, and blue showing down-regulation. TNF: Tumor necrosis factor; IL6: Interleukin 6.

Figure 8

Gene ontology enrichment analysis of gene targets of differentially expressed miRNAs. A: Vascular cell adhesion molecule-1 group vs control group; B: Tumor necrosis factor α group vs control group; C: Interleukin 6 group vs control group. The X axis shows the number of differently expressed genes (their square root value), and the Y axis shows GO terms. All GO terms are grouped into three ontologies: Blue indicates biological process, brown indicates cellular components, and red indicates molecular function

Figure 9

Pathway enrichment analysis of gene targets of differentially expressed miRNAs. A: Comparison of the top 20 enriched pathway terms between vascular cell adhesion molecule-1 (VCAM-1) group and control group; B: Comparison of the Kyoto Encyclopedia of Genes and Genomes (KEGG) classification between VCAM-1 group and control group; C: Comparison of the top 20 enriched pathway terms between tumor necrosis factor (TNF) α group and control group; D: Comparison of the KEGG classification between TNFα group and control group; E: Comparison of the top 20 enriched pathway terms between interleukin 6 (IL6) group and control group; F: Comparison of the KEGG classification between IL6 group and control group. A, C, and E: The top 20 enriched pathway terms displayed as scatterplots. The rich factor is the ratio of target gene numbers annotated in this pathway term to all gene numbers annotated in this pathway term. The greater the rich factor, the greater the degree of enrichment. The Q-value is the corrected P-value and ranges from 0-1; the lower the Q-value, the greater the level of enrichment; B, D, and F: The X axis shows the number of target genes, and the Y axis shows the second KEGG pathway terms. The first pathway terms are indicated using different colors. The second pathway terms are subgroups of the first pathway terms and are grouped together on the X axis on the right side.

Figure 10

Gene ontology enrichment analysis of angiogenesis-related genes. A: The number of angiogenesis-related genes in the vascular cell adhesion molecule-1 (VCAM-1) group; B: The proportion of angiogenesis-related genes in the VCAM-1 group; C: The number of angiogenesis-related genes in the tumor necrosis factor (TNF) α group; D: The proportion of angiogenesis-related genes in the TNFα group; E: The number of angiogenesis gene distributions in the interleukin 6 (IL6) group; F: The proportion of angiogenesis-related genes in the IL6 group; G: Comparison of the number of angiogenesis-related genes in the three groups to that of the control group.

Figure 13

Regulatory mechanism of the VEGF signal pathway in different groups. A: Vascular cell adhesion molecule-1 group; B: Tumor necrosis factor α group; C: Interleukin 6 group. Up-regulated genes are marked with red borders and down-regulated genes with green borders. Unchanged genes are marked with black borders.

Regulatory mechanism of the PI3K-AKT signal pathway in different groups. A: Vascular cell adhesion molecule-1 group; B: Tumor necrosis factor α group; C: Interleukin 6 group. Up-regulated genes are marked with red borders and down-regulated genes with green borders. Unchanged genes are marked with black borders.

Regulatory mechanism of the MAPK signal pathway in different groups. A: Vascular cell adhesion molecule-1 group; B: Tumor necrosis factor α group; C: Interleukin 6 group. Up-regulated genes are marked with red borders and down-regulated genes with green borders. Unchanged genes are marked with black borders.

MSCs-exo promote endothelial cell angiogenesis

The ability of MSCs-exo to enhance tube formation was assessed using a Matrigel assay. MSCs-exo, on average, caused an increase in tube formation and branching (P < 0.05; Figure 14) compared to the untreated control group. The angiogenesis effect of MSCs-exo stimulated with TNFα and IL6 was lower than that cultured without TNFα or IL6. This finding confirmed that MSCs-exo can promote angiogenesis (Figure 14).

Figure 14

Tube formation in endothelial cells treated with mesenchymal stem cell exosomes. A: Micrograph showing human umbilical vein endothelial cells (HUVECs) cultured on Matrigel-coated plates in medium with phosphate buffered saline (control); B: Micrograph showing HUVECs cultured on Matrigel-coated plates in medium with mesenchymal stem cell exosomes (MSC-exo); C: Micrograph showing HUVECs cultured on Matrigel-coated plates in medium with MSCs-exo stimulated with tumor necrosis factor α; D: Micrograph showing HUVECs cultured on Matrigel-coated plates in medium with MSCs-exo stimulated with interleukin 6; E: The number of tubes formed in each group; F: The number of branching points in each group.

MSCs-exo stimulated with IL6 inhibit PI3K-AKT signaling pathway activation in endothelial cells

We observed a decline in the expression of phosphorylated AKT (pAKT) in endothelial cells after MSCs-exo were stimulated with IL6, although MSC-EV treatment had no effect on the expression of their unphosphorylated forms (Figure 15).

Figure 15

Western blot analysis. Western blot was performed to detect the expression of the indicated proteins in endothelial cells treated with phosphate buffered saline (control), mesenchymal stem cell exosomes (MSCs-exo), MSCs-exoTNFα (stimulated with tumor necrosis factor α), MSCs-exoIL6 (stimulated with interleukin 6). GAPDH was used as an internal loading control. TNF: Tumor necrosis factor; IL6: Interleukin 6.

DISCUSSION

MSCs have been used to treat cardiovascular diseases (CVDs) and represent a promising cell-based therapy for regenerative medicine and the treatment of inflammatory and autoimmune diseases[22]. Their treatment efficacy hinges on their ability to alter disease-specific pathways via secreted miRNAs, so it is imperative to understand how disease environments, which often are inflammatory, can impact secreted miRNAs and thus potentially their treatment efficacy. MSCs-exosomes are used for treating CVDs such as acute myocardial infarction, stroke, pulmonary hypertension, and septic cardiomyopathy[23]. Biological properties of MSCs-exo have rendered them as a new strategy for wound regeneration and ischemic disease[24-26]. MSCs-exo exert an anti-inflammatory effect on T and B lymphocytes independently of MSCs priming. The potential therapeutic effects also were demonstrated in inflammatory arthritis[27].

Ischemic diseases, trauma, and immunological diseases are all accompanied by inflammatory reactions, and a large number of inflammatory cytokines including VCAM-1, TNFα, and IL6 are involved in the progression of these diseases. T cells are activated dependent on VCAM-1 interactions[28]. VCAM-1 plays a “backup” role in hASC contact-dependent immune suppression[29]. TNFα is a multifunctional cytokine that acts as a central biological mediator for critical immune functions, including inflammation, infection, and antitumor responses[30]. IL6 plays an important role in the inflammatory response following hypoxic-ischemic encephalopathy[31]. Therefore, the biological properties of MSCs-exos and their therapeutic effects need to be studied together with the inflammatory factors. A large amount of evidence suggests that the effect of MSCs-exo therapeutics will be affected by the inflammatory environment with regard to most of CVDs and ischemic diseases[32-40].

Some previous reports showed[41] the effect of stimulation with cytokines interferon γ and TNFα on adipose MSCs (AMSCs). Pro-inflammatory stimuli could enhance the immunosuppressive functions of AMSC-derived exosomes. There was an increase in the expression of miRNAs (miR-34a-5p, miR21, and miR146a-5p) in exosomes produced by pre-activated AMSCs compared to those released by untreated cells.

The present study also found that the inflammatory cytokines VCAM-1, TNFα, and IL6 impact the size and morphology of MSCs-exo as well as the diversity of miRNAs they can produce, especially miRNAs impacting angiogenesis. According to GO enrichment analysis, miRNAs in exosomes exposed to inflammatory cytokines, compared to controls, had a different regulatory effect on cellular proliferation and differentiation, molecular signal transduction, immunosuppressive functions, angiogenesis and so on.

Some observed effects suggested that inflammatory cytokines impaired the ability of MSCs-exo to promote angiogenesis. For example, the TNFα and IL6 groups exhibited decreased numbers of angiogenesis-related miRNAs, such as miR-196a-5p, miR-17-5p, miR-146b-5p, miR-21-3p, and miR-320. The same groups also had downregulated angiogenesis-related signaling pathways, such as PI3K-AKT, MAPK, and VEGF. However, other effects suggested that inflammatory cytokines may promote the ability of MSCs-exo to encourage angiogenesis. Exosomes contained hsa-mir-4488, hsa-mir-671-5p, and hsa-mir-4446-3p after VCAM-1 stimulation, hsa-mir-4488, hsa-mir-671-5p, and hsa-miR-497-5p after TNFα stimulation, and hsa-mir-4488, hsa-miR-145-5p, and hsa-miR-1260a after IL6 stimulation, all of which promote angiogenesis. More specifically, hsa-miR-671-5p encourages NM_006500.2 to produce the downstream product VEGFb, which activates the VEGF pathway and thereby promotes angiogenesis. Hsa-miR-671-5p also encourages NM_001773.2 to produce CD34, which in turn activates the PI3K-AKT pathway, this promoting cell proliferation and angiogenesis. It is unclear what the ultimate results of these competing effects on angiogenesis would be.

Our study demonstrated that MSCs-exo perhaps induced HUVECs to form capillary-like structures in vitro. The effect of MSCs-exo in promoting angiogenesis would be reduced when the stem cells were subjected to TNFα and IL6 stimulation. Besides endothelial cell angiogenesis-related molecular expression, functional characteristics such as the PI3K-AKT signaling pathway may be down-regulated in MSCs-exo that were stimulated with IL6.

The major limitation of this study is that it was conducted in vitro. Further research needs to be conducted in vivo using an animal model that more closely mirrors the inflammatory environment to which MSCs are exposed when administered for treatment in humans.

In conclusion, inflammatory cytokines may lead to changes in exosomal miRNAs that abnormally impact cellular components, molecular function, and biological processes. Further in vivo research needs to be conducted to explore how the treatment efficacy of MSCs is impacted by these inflammatory-induced changes in exosomes and their miRNAs.

ARTICLE HIGHLIGHTS

Research background

Stem cell transplantation has been developing rapidly and has resulted in breakthroughs for the treatment of various diseases.

Research motivation

Treatments utilizing stems cells often require stem cells to be exposed to inflammatory environments, such as vascular cell adhesion molecule-1, tumor necrosis factor α (TNFα), and interleukin 6 (IL6). Stem cell-derived exosomes are especially important in producing miRNAs that impact angiogenesis.

Research objectives

MicroRNAs (miRNAs) are RNAs 0-20 nucleotides in length, which are derived from hairpin-like precursor miRNAs. They acts as important regulators of mRNA expression. It has been reported that miRNAs play critical roles in some cells and have the potential as diagnostic and therapeutic biomarkers.

Research methods

The morphology and quantity of mesenchymal stem cell (MSC) exosomes (MSCs-exo) are influenced by different inflammatory cytokine environments.

Research results

The morphology and quantity of each group of MSC exosomes were observed and measured. The miRNAs in MSCs-exo were sequenced. Differential expression of miRNAs and their target genes as well as the related regulatory mechanisms were researched.

Research conclusions

TNFα and IL6 may influence the expression of miRNAs that down-regulate the PI3K-AKT, MAPK, and VEGF signaling pathways; particularly, IL6 significantly down-regulates the PI3K-AKT signaling pathway.

Research perspectives

Overall, inflammatory cytokines may lead to changes in exosomal miRNAs that abnormally impact cellular components, molecular function, and biological process, particularly angiogenesis.

ACKNOWLEDGEMENTS

The authors would like to thank all members of the Jinan University Biomedical Translational Research Institute Laboratory who provided us with critical comments and support.

Table 3

Comparison of differentially expressed angiogenesis-related miRNAs between the TNFα group and control group

miRNA id	Count (control)	Count (TNFα)	TPM (control)	TPM (TNFα)	Log2 ratio (TNFα/ control)	Up/ down-regulation (TNFα/ control)	P	FDR
hsa-miR-196a-5p	26485	8359	1616.96	403.05	-2.004253263	Down	0	0
hsa-miR-320e	161	46	9.83	2.22	-2.14663174	Down	5.95E-23	6.85E-22

P-values were corrected. TPM: Transcripts per kilobase million; FDR: False discovery rate; TNF: Tumor necrosis factor.