Characterization of exosomes
The morphological feature of exosomes was observed by transmission electron microscope. As shown in Fig. 1A, most of the photo-captured round-shaped microvesicles are of a size of about 100 nm in diameter with the membrane bounded. The size distribution of the serum exosomes is shown in Fig. 1B. The protein markers of exosomes (TSG101) are shown in Fig. 1C. The results indicated that exosomes can be successfully isolated from serum using this protocol with quality adequate for subsequent experiments.
Sequence analysis of small RNAs
Thirteen sRNA libraries were generated from serum exosomes via high-throughput sequencing. We used Fast-QC software (http://www.bioinformatics. babraham.ac.uk/projects/fastqc/)for the overall evaluation of the quality of sequencing data and used BWA algorithm for small RNA mapping. There are 15548736, 13804825, 12383415, 16332234, 15327209, and 16970328 total reads in Ctr1, Ctr 2, CF1, CF2, PF1, and PF2 sRNA libraries, respectively (Table 2). The corresponding clean reads of 2654371(17.07%), 2911044(21.09%), 643916(5.2%), 5843973(35.8%), 2587012(16.9%) were respectively obtained (Table 2) after the removal of the low-quality reads, including 3'adapter null, insert null, 5'adapter contaminants, reads smaller than 18nt, and reads containing ploy A. The miRNA rates in total RNA were 16.03%, 2.01%, 1.66%, 2.17%, 7.17%, and 2.09%, respectively.
Table 1
Basic statistical information on the quality of the reads
Sample
|
Ctr1
|
Ctr2
|
CF1
|
CF2
|
PF1
|
PF2
|
Raw reads
|
15548736
|
13804825
|
12383415
|
16332234
|
15327209
|
16970328
|
Clean reads
|
2654371
|
2911044
|
643916
|
5843973
|
2587012
|
4870855
|
Clean Rate(%)
|
17.07
|
21.09
|
5.20
|
35.80
|
16.90
|
28.70
|
Q20(%)
|
95.65
|
94.25
|
93.21
|
94.72
|
94.15
|
96.4
|
Q30(%)
|
90.78
|
88.42
|
86.61
|
89.02
|
88.28
|
92.46
|
GC(%)
|
43.7
|
43.17
|
42.48
|
44.37
|
41.55
|
38.86
|
tRNA(%)
|
10.81
|
3.54
|
13.66
|
38.01
|
9.78
|
7.65
|
rRNA(%)
|
46.4
|
70.41
|
38.14
|
31.72
|
36.19
|
47.25
|
miRNA(%)
|
16.03
|
2.01
|
1.66
|
2.17
|
7.17
|
2.09
|
Differentially expressed miRNAs
We applied DESeq2 algorithm to filter the differentially expressed genes. An miRNA candidate is considered to be differentially expressed if the level between the two experiment groups shows a statistically significant difference (p < 0.05) of at least two folds. Compared with CF group, 54 types of miRNAs are down-regulated, and 154 are up-regulated in PF group. (Fig. 2a). The differently expressed miRNA profiles (Fig. 2B) were obtained from comparative analyses of the exosomes in the three groups (control. CF, and PF) according to the value of log2FC > 2 miRNA whose expression level was respectively highest or lowest. As shown in Table S3, the top five most down-regulated miRNAs in the patient group were miR-135a-5p, miR-196a-5p, miR-34b-3p, miR-100-5p, and miR-125b-5p, while the top five most up-regulated miRNAs were miR-4454, miR-4286, miR-5100, miR-7977, and miR-624-5p.
Table 2
Difference miRNA for down-regulation or up-regulation (Top 5)
|
down-regulation miRNA
|
up-regulation miRNA
|
Number
|
miRNA
|
log2FC
|
miRNA
|
log2FC
|
1
|
miR-135a-5p
|
-7.73
|
miR-4454
|
7.72
|
2
|
miR-196a-5p
|
-6.18
|
miR-4286
|
7.43
|
3
|
miR-34b-3p
|
-6.01
|
miR-5100
|
5.95
|
4
|
miR-100-5p
|
-5.49
|
miR-7977
|
5.75
|
5
|
miR-449b-5p
|
-5.14
|
miR-624-5p
|
5.39
|
Validation miRNA expression by q-PCR
To further validate our miRNA profiling results, we screened 10 patients and 10 healthy people for validation of miRNA expression levels.We found that the expression trends of all miRNAs determined by q-PCR were identical to those obtained from the RNA sequencing (Fig. 3). Compared with the control group, the expression levels of miR-196a-5p and miR-449b-5p decreased, whereas those of miR-4454 and miR-5100 increased, both to a significant degree (p < 0.05). These observations suggest that miR-196a-5p, miR-449b-5p, miR-4454, and miR-5100 were involved in the pathogenesis of EMS and that they may serve as potential biomarkers of EMS.
Table 3
The miRNA primers sequence for q-PCR
miRNA Name
|
Forward primer
|
Reverse primer
|
U6
|
GCTTCGGCAGCACATATACTAAAAT
|
CGCTTCACGAATTTGCGTGTCAT
|
miR-135a-5p
|
ACACTCGAGCTGGGTATGGCTGTTTATTCCT
|
TGGTGTCGTGGAGTCGGC
|
miR-449b-5p
|
ACACTCCAGCTGGGAGGCAGTGAATTGTTA
|
TGGTGTCGTGGAGTCGGC
|
miR-654-3p
|
ACACTCCAGCTGGGTATGTCTGCTGAGCAT
|
TGGTGTCGTGGAGTCGGC
|
miR-4449
|
ACACTCCAGCTGGGCGTCCCGGTGCTGCGC
|
TGGTGTCGTGGAGTCGGC
|
hsa-miR-5100
|
ACACTCCAGCTGGGTTCAGATCGCAGCGGT
|
TGGTGTCGTGGAGTCGGC
|
hsa-miR-223-3p
|
ACACTCCAGCTGGGTGTCAGTTAGTCAAAT
|
TGGTGTCGTGGAGTCGGC
|
ROC curve analysis of exosomal miRNA as diagnostic biomarkers for EMS
We subsequently used ROC curve analysis to assess the potential of the 6 miRNAs (Table 3) as EMS biomarkers. The AUC of miR-135a-5p, miR-196a-5p, miR-449b-5p, miR-4454, miR-4286, and miR-5100 were 0.53 (95% CI, 0.400–0.800), 0.570 (95% CI, 0.700 − 0.600), 0.680 (95% CI 0.600-1.000), 0.956 (95% CI, 0.800-1.000), 0.520 (95% CI, 1.000-0.200), 0.900 (95% CI, 0.800–0.900), respectively, suggesting that miR-4454 and miR-5100 from serum exosomes hold a high potential as biomarkers of EMS.