Research Article

Isolation of Serum Exosomes and Correlation of its Contents with Acute Ischemic Stroke

Wencheng Chen*, Pei Yu, Yanfang Pang, Mengran Han, Huixiong Yuan , Yanxin Wang

Department of Clinical Laboratory, The Affiliated Hospital of Youjiang Medical University for Nationalities, Guangxi, China

*Corresponding author: Wencheng Chen, Department of Clinical Laboratory, The Affiliated Hospital of Youjiang Medical University for Nationalities, No. 18 Zhongshan Second Road, Youjiang District, Baise, Guangxi, China, 533000. Tel: +776-2825603; E-mail: chwch3268@sina.com

Abstract

Background: Erythrocyte deformability is one of the pathophysiological changes of high-risk factors in stroke. It mainly affects blood viscosity and fluidity in the occurrence, development and outcome of the disease. Exosomes are a new type of biological activity test target, and it mainly contains miRNA, but its effect on stroke is still not clear.

Aims: To determine the expression of exosomal miR-150-5p in acute ischemic stroke patients and to explore the potential value of this substance in predicting the risk of acute ischemic stroke and providing new ideas for clinical diagnosis.

Methods: The blood samples from acute ischemic stroke (AIS) patients and normal controls with age and sex-matched were collected. The exosomes were isolated from the blood samples, which were confirmed by electron microscopy and western blot with the specific exosomes biomarker CD63 and HSP70. Real-time fluorescent quantitative polymerase chain reaction(qRT-PCR) was used to determine the expression of serum exosomal miR-150-5p. The Areas Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) curve were constructed to evaluate the diagnostic accuracy of miR-150-5p in AIS. Bioinformatics methods predict the target genes of miR-150-5p through an online database.

Results: The expression levels of exosomal miR-150-5p were significantly increased in AIS patients compared with that of the control group (P<0.001). The exosomal miR-150-5p in AIS groups were monitored at 24, 48 and 72 hours after the onset of acute stroke were not statistically significant (P>0.05). ROC analysis showed high AUC values for miR-150-5p as compared to those in healthy controls was 0.883(P<0.001). Bioinformatics analysis of miR-150-5p predictable target genes, it turns out that miR-150-5p may be through EGR2 and PLP2 involved in the development of stroke disease, as well as erythrocyte membrane protein-related genes SLC4A1 and SPTB also belong to the miR-150-5p prediction targets.

Conclusion: The expression of exosomal miR-150-5p is relatively high in patients with acute ischemic stroke, and it may have a certain potential for early disease diagnosis.

Key words: Acute Ischemic Stroke; Exosomes; Non-Coding RNA; Erythrocyte.

Introduction

Ischemic strokes usually happen when blood vessels are blocked by vascular deposits of thrombus in blood vessels supplying the brain. Without timely elimination of this blockage, brain tissue ischemia may lead to a permanent neurological deficit or brain death. Approximately 84% deaths are caused by ischemic strokes [1]. Stroke diagnosis currently relies on imaging techniques. However, it might be difficult for physicians to make objective judgments when symptoms are mild or difficult. In recent years, epigenetics and gene sequence analysis have been increasingly used as adjuncts in the diagnosis of cardiovascular disease and cancer. MiRNA, in particular, has been indicated to be associated with several risk factors for stroke, including hypertension, diabetes, atherosclerosis, etc [2]. Exosomes are small membrane-bound vesicles secreted by most types of cells. They contain a variety of functional substances that can be used for diagnostic and therapeutic purposes, such as mRNAs and miRNAs [3].

As thrombus persists during ischemic stroke, changes in the erythrocyte membrane and the resulting material may trigger other possible structural and biochemical changes such as red blood cells in the circulatory system [4]. For patients with acute cerebral circulation disorders, their blood viscosity increases, and erythrocyte aggregates, erythrocyte deformability reduces, and plasma fibrinogen levels elevate [5]. As for patients with thromboembolic ischemic stroke, scanning electron microscope observation indicates 92% abnormal erythrocytes in the smear [6]. The change of erythrocyte deformation capacity affects both the oxygen transport capacity of erythrocytes and the survival rate of circulating erythrocytes since deformability of erythrocytes plays a crucial role in the circulation of erythrocytes through the microvasculature system and splenic sinusoids. Besides, reduced cellular deformability is closely related to stroke [7]. Researchers found that sustained inhibition of miR-150 is essential for normal final erythroid development. It is also validated the 4.1R gene is a new target of the miR-150 during terminal erythropoiesis. Its enrichment ensures mechanical stability and deformability of the membrane [8]. Therefore, in this study, we extracted and identified serum exosomes detecting exosomal miR-150-5p expression, to investigate whether exosomal miR-150-5p is altered in patients with acute stroke and its diagnostic value as a new biomarker.

Material and Methods

Clinical sample collection

Forty-two patients were diagnosed with acute ischemic stroke during hospitalization in the Neurology Department of the Affiliated Hospital of Youjiang Medical University for Nationalities between May 2020 and November 2020. The severity of stroke was evaluated by the National Institutes of Health Stroke Scale (NIHSS) by experienced neurologists. The Control group were forty-two healthy people who visited the hospital during the same period for routine physical examinations and were denied a history of strokes. Inclusion criteria for patients1. Age ≥ 45 years old with no previous history of stroke; 2. The first episode with a duration of no more than 72 hours; 3. Diagnosis through examination by the clinical neurologist and confirmed by MR and CT scans. Exclusion criteria: recurrent stroke or duration of stroke attack over 72 hours, renal or hepatic failure, acute infectious disease, tumor, hematological diseases, and patients who are unable to cooperate with physical examinations. The serum samples were collected from peripheral blood via centrifugation and stored at −80°C. This study was approved by the Ethics Committee of the Affiliated Hospital of Youjiang Medical University for Nationalities (Grant No. YYFY-LL-2019-001). Written consent was obtained from each participant. By strictly following the principles of the Declaration of Helsinki. The clinical characteristics of acute stroke patients and controls were in table 1.

 

Parameter

AIS(N=42)

Control (N=42)

Statistics

P value

Sex (male/female)

28/14

23/19

X2=1.248

0.264

Age (year)

60.05±8.36

57.07±6.56

t=1.815

0.073

Hypertension (%)

25 (59.5%)

10 (23.8%)

X2=11.02

0.001

Diabetes mellitus (%)

9 (21.4%)

6 (14.3%)

X2=0.730

0.393

Smoke (%)

15 (35.7%)

8 (19.0%)

X2=2.934

0.087

Drink (%)

21 (50.0%)

18 (42.9%)

X2=0.431

0.512

Hyperlipidemia (%)

20 (47.6%)

9 (21.4%)

X2=6.372

0.012

AST (U/L)

21.28±5.21

20.15±4.69

t=1.039

0.302

ALT (U/L)

18.55±6.85

18.42±5.58

t=0.096

0.924

TG (mmol/L)

1.42(1.06,1.99)

0.92(0.79,1.33)

z=3.619

<0.001

TC (mmol/L)

4.90±1.25

4.57±0.77

t=1.434

0.156

HDL-C (mmol/L)

1.10±0.35

1.41(1.14,1.52)

z=-4.017

<0.001

LDL-C (mmol/L)

3.54±1.27

3.04±0.66

t=2.251

0.027

Glu (mmol/L)

4.81±0.89

4.80±0.46

t=0.055

0.956

UA (umol/L)

325.98±88.97

319.88±68.00

t=0.353

0.725

WBC (x109/L)

7.49(6.01,9.55)

6.79±1.66

z=2.178

0.029

RBC(x1012/L)

4.70±0.57

5.26±0.60

t=-4.455

<0.001

Hb(g/L)

138.62±14.41

144.26±15.21

t=-1.745

0.085

HCT(%)

42.04±4.75

44.45±3.85

t=-2.559

0.012

PLT(x109/L)

238.29±70.39

268.48±63.64

t=-2.062

0.042

NIHSS

1(0,4)

-

-

-

Table 1: Baseline clinical characteristics of the AIS patients and normal controls. 

 

Serum collection and exosomes isolation

Collect 5mL of whole blood in a clean centrifuge tube and leave it for the clot to separate. Centrifuge at 4000×g,4℃ for 10min, and remove the supernatant. Centrifuge again at 1500×g,4℃ for 10min, and separate into 1.5ml enzyme-free centrifuge tubes. Exosomes were extracted from serum using the Exosome Precipitation Reagent Kit (EZB-exo101, EZBioscience, USA) following the manufacturer's protocol. One portion of the extract was used for total RNA extraction and another portion was used for exosome identification.

Exosome identification

The extracted exosome was resuspended in enzyme-free water and dropped on a copper mesh with carbon membrane support for 3-5 min, then removed the excess liquid with filter paper; 2% phosphotungstic acid was dropped on a copper mesh with carbon membrane support for 2-3 min, and the excess liquid was then removed with filter paper. Dried at room temperature, observed under a transmission electron microscope (TEM, HT7700 Hitachi, Japan) and collected images for analysis.

Intact receptors and antigens were preserved on the surface of exosomes, such as CD63, CD9, Tsg101, Alix, HSP70, etc [9] are common key test indicators. Skimmed milk was used to seal the blotting membranes and was incubated overnight with CD63 antibodies (1:1000, Abcam, UK), HSP70(1:1000, Abcam, UK). Applied HRP-conjugated Goat Anti-Rabbit Secondary Antibody (1:5000CSTUSA) to CD63, and applied HRP-conjugated Goat Anti-Mouse Secondary Antibody to HSP70 for 1 hour (1: 5000, Sanying, Wuhan, China), applied imaging scans to PVDF membranes through the gel imaging system.

qRT-PCR for exosomal miR-150-5p expression

Exosome RNA Purification kit (EZBioscience, EZB-Exo-RN1, USA) was to acquire RNA. Reverse transcription of RNA using miDETECT A Track miRNA qRT-PCR Starter Kit (Ribo, Guangzhou, China) under the following conditions: mixing with a reverse transcription reaction system. The reaction was performed at 42°C for 1 hour, followed by incubation at 72°C for 10 minutes to obtain cDNA, SYBR Green was used as a fluorescent molecule using a three-step process: 95°C, 10 minutes, 1 cycle; 95°C, 2 seconds, 60°C, 20 seconds, 70°C, 10 seconds, 40 cycles; dissolution curve generated. External reference Cel-miR-39-3p (miDETECTTM miRNA External Control, Guangzhou Yuebo) was used for standardized processing. The relative expression of miRNA was quantified through the 2-△△Ct method.

Statistics

All data were processed with SPSS 24.0. Student’s t-tests of the two independent samples were used to compare differences between the two groups. One-way ANOVA was used to analyze the difference between more than two groups. If measures did not conform to a normal distribution, a non-parametric test was performed. Results were considered statistically significant when the p-value was less than 0.05.

Result

General clinical information and baseline information sheet

The ratios and age of men and women indicating no statistical significance. In terms of high-risk factors for stroke, the differences between the two groups show no statistical significance in the history of diabetes, smoking, drinking, etc. However, the difference in hypertension and hyperlipidemia, the AIS group demonstrates a higher incidence rate (P < 0.05). TG, HDL-C, LDL-C, WBC, and HCT were higher in the AIS group, while RBC and PLT counts are higher in the control group (P <0.05). Exosomal miR-150-5p and NIHSS scores show no correlation.

Exosome identification

Electron microscopy shows oval or round membranous vesicles, similar in shape and size, with a diameter between 30-200nm. Some of the vesicles are fused, and vesicles are rare in some exosomes [10] (Figure 1). CD63 and HSP70 protein expression can be tested in the exosomes extracted from both the control group and the AIS group (Figure 2).

 

Figure 1: Exosomes observed under transmission electron microscope.

 

 

Figure 2: Western Blot analysis of exosome-specific marker proteins in AIS and control group, detection of exosome-specific marker proteins CD63 and HSP70.

 

Expression trends of exosomal miR-150-5p in patients with acute ischemic stroke

The Comparison of the relative expression levels of exosomal miR-150-5p in the two groups (Figure 3). With cel- miR-39-3p as a control, exosomal miR-150-5p relative expression in the disease group was 1.79 ± 0.45, higher than that of the control group, 1.06±0.38(P<0.001). The exosomal miR-150-5p in AIS groups were monitored at 24, 48 and 72 hours (Figure 4) after the onset of acute stroke were not statistically significant(P>0.05).

 

Figure 3: Relative expression level of serum exosome miR-150 in AIS group and control group: AIS group n = 42; control group n = 42, T=8, ***P<0.001.

 

 

Figure 4: At different time points after the occurrence of stroke, the expression level of serum exosome miR-150 in AIS patients was N=42, NS, no difference, Chi-Square=2.925, P=0.232, P>0.05, Kruskal-Wallis H test.

ROC curve analysis of exosomal miR-150-5p

Receiver Operating Characteristic curve (ROC) curves are suitable for the analysis and evaluation of diagnostic tests, especially the evaluation and comparison of clinical diagnostic tests. Performed ROC curve analysis to exosomal miR-150-5p (Figure 5), the area under the curve was 0.883 (95%CI: 0.810-0.956, P<0.01), indicating a certain diagnostic value.

 

Figure 5: Diagnostic value of serum exosome miR-150 in acute stroke: ROC curve was generated based on the relative level of serum exosome miR-150 to distinguish patients with AIS (n = 42) from normal controls (n = 42), The cured area is 0.883 (95% confidence interval = 0.81-0.96).

 

Target gene prediction

The results of miR-150-5p target genes are predicted based on four commonly used online tools: DIANA predicts 915 target genes for miR-150-5p; miRanda predicts 525; Pictar predicts 264; TargetScan predicted 363. Finally, eight co-expressed target genes (HILPDA, FOXD3, EGR2, PLP2, ELOVL3, ZNF-335, SHISA4, STX5) were analyzed online using Draw Venn Diagrams (Figure 6). The literature found that PLP216EGR2 [11,12] is closely related to the occurrence and development of stroke, and Erythrocyte membrane protein-related genes SLC4A1 and SPTB are also miR-150-5p targets [13,14].

 

Figure 6: miR-150 target gene prediction Wayne diagram: prediction of miR-150 target gene results based on four commonly used online tools: DIANAmiRandaPictarTargetScan.

 

We used DAVID to obtain target genes ontology (GO) information and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis to annotate potential functions, P<0.05 as a significant level of significance. The GO information of predicted target genes shows that most genes have binding functions with proteins in terms of molecular function, but the predicted target genes are not enriched in related signal pathways (Table 2).

 

Category

Term

Count

Cellular Components

endoplasmic reticulum membrane

3

Molecular Function

protein binding

6

Table 2: The numbers of genes enriched in each GO category. 

 

Discussion

Circulating miRNA is associated with exosomal miRNA

The miRNAs concentrated in the brain play an important role in the pathophysiological mechanism of stroke. They are stored in the blood and cerebrospinal fluid and show stability in long-term storage [15,16] Notably, circulating miRNAs are primarily derived from exosomal miRNAs. Although circulating miRNAs and exosomal miRNAs are both potential serum markers of stroke, the very same circulating miRNA or exosomal miRNAs during the same disease may be different. Chen F et al [17] studied the severity idiosyncrasy of the change of serum and exosomal miR-126 levels to cerebral ischemia and found no correlation between blood serum and exosomal miR-126 level. Circulating miRNAs are usually present in exosomes and are sometimes bound by certain carrier proteins to maintain stable [18,19]. This stability is considered a result of the fact that exosomes wrap the miRNA and protect it from the interference of external RNA enzymes [20]. During ischemic stroke, it will result in oxidation of red blood cells and changes in protein hydrolysis and therefore lead to changes in cellular rheology and inflammatory processes. Regulated miRNA expression is crucial for erythrocyte chromatin condensation and denucleated terminal red lineage differentiation [21].

Changes of miR-150-5p in patients with acute ischemic stroke

In this study, we found out the exosomal miR-150-5p relative expression levels of the control group and the acute disease group were significantly different, with the relative expression of the disease group higher than that of the control group. Monitoring of 24h, 48h, and 72h after onset shows that the difference in the relative expression level of exosomal miR-150-5p was not statistically significant.

Bioinformatics prediction analysis of miR-150 target genes, which may be involved in the development of stroke disease through EGR2 and PLP2, as well as the erythrocyte membrane protein-related genes SLC4A1 and SPTB, also belong to the miR-150 prediction targets. predicted target genes shows that most genes have binding functions with proteins in terms of molecular function, but are not enriched in related signal pathways. Research shows that regulation of miR-451 and miR-150 expression may be an effective alternative for stimulatory cytokines to differentiate CD133+ into erythrocyte-like spectrum [22]. Scherrer et al found that among patients suffering from ischemic stroke, miR-150-5p was highly correlated with mortality rate within 90 days [23]. In the rat model that suffered from cerebral ischemia, upregulation of miR-150 expression reduces vascularity in the infarct border zone of rats after MCAO and decreases tube formation, proliferation, and migration of cerebral microvascular endothelial cells [24]. Studies involving post-stroke blood-brain barrier disruption mechanisms during edema and hemorrhagic transformation indicated that miR-150 can affect the blood-brain barrier permeability after permanent middle cerebral artery occlusion in rats [25].

Selection of internal and external references in the study of exosomal miRNA

Meanwhile, cel-miR-39 was chosen as an internal reference. A certain amount of standard was added during the lifting process to normalize the relative expression levels of exosomal miR-150-5p. U6 was usually selected as an internal reference gene in non-coding RNAs. It is stable, abundant, and highly conserved, and prevails in all eukaryotic cells. Considering repeated freeze-thawing often occurs during experiments, U6 is not suitable to be the research reference of circulating miRNA, since longer RNAs are more likely to be degraded by RNase. And U6 is significantly longer than miRNA. Also, U6 contains a large number of nucleotides that may be degraded by RNase in blood after separation from carrier proteins during freezing and thawing [26]. Some researchers prefer to use circulating miRNAs such as miR-16 with a stable expression as an internal reference [27]. However, Ostenfeld considered that miR-16 is down-regulated in rectal cancer tissues and can be used as a biomarker of prognosis [28]. Given the uncertainty of non-coding RNA expression in blood, exogenous miRNAs of the same amount with internal reference were added into serum or plasma samples. Common exogenous miRNAs include cel-miR-39 of the showy primer nematode and lin-4. These exogenous miRNAs have no human homologous gene sequences. A certain amount of exogenous miRNAs were added during RNA extraction, and experience extraction, reverse transcription, and PCR quantification together with the target miRNAs to be tested. This method eliminates bias in the examination and solidates the results. However, it also cumbersome the experimental procedures, as well as clinical applications. Yet, manual adulterated miRNAs are free heterologous miRNA molecules that are not affected by changes in blood composition and do not reflect the changes in circulating miRNA expression caused by changes in the internal environment of the human body, and therefore cannot improve the accuracy of analysis [29].

Limitations and Outlook

Some limitations still exist in this study: small sample size limits the overall results, therefore, it is necessary to expand the sample amount to better prove that the upregulation of relative expression level of exosomal miR-150-5p in acute stroke patients is not a coincidence; since most of the subjects come from the Zhuang ethnic group in Guangxi Zhuang Autonomous Region, the geographical restriction blocks the knowledge of difference among regions and ethnic groups; there is no uniformity in the selection of blood reference genes for exosomal miRNA detection; many functions of the organism do not rely on a single regulator mechanism, functional validation of predicted target genes involved competing RNAs is not fully understood; we didn’t trace the origin of exosomes, whether there are differences or correlations between circulating blood miRNA-150-5p and exosomal miRNA-150-5p could be part of the next study.

Acknowledgment

This study was funded by the Scientific Research and Technology Open Plan Project of Baise (grant numbers 20202501).

References

  1. Goldstein LB, Adams R, Becker K, Furberg CD, Gorelick PB, et al. (2001) Primary prevention of ischemic stroke: A statement for healthcare professionals from the Stroke Council of the American Heart Association. Stroke 32: 280-99.
  2. Schober A, Weber C (2016) Mechanisms of MicroRNAs in Atherosclerosis. Annu Rev Pathol 11: 583-616.
  3. Zhang J, Li S, Li L, Li M, Guo C, et al. (2015) Exosome and exosomal microRNA: trafficking, sorting, and function. Genomics Proteomics Bioinformatics 13: 17-24.
  4. Pretorius E, Lipinski B (2013) Thromboembolic ischemic stroke changes red blood cell morphology. Cardiovasc Pathol 22: 241-242.
  5. Ionova VG (2003) State of the Rheological Properties of Blood in the Dynamics of Development of Acute Disturbances of Cerebral Circulation in Atherosclerosis Patients. Journal of Engineering Physics & Thermophysics 76: 672-680.
  6. Swanepoel AC, Pretorius E (2012) Scanning electron microscopy analysis of erythrocytes in thromboembolic ischemic stroke. Int J Lab Hematol 34: 185-191.
  7. Tikhomirova IA, Oslyakova AO, Mikhailova SG (2011) Microcirculation and blood rheology in patients with cerebrovascular disorders. Clin Hemorheol Microcirc 49: 295-305.
  8. Sun Z, Wang Y, Han X, Zhao X, Peng Y, et al. (2015) miR-150 inhibits terminal erythroid proliferation and differentiation. Oncotarget 6: 43033-43047.
  9. Kalluri R, LeBleu VS (2020) The biology, function, and biomedical applications of exosomes. Science 367: 6478.
  10. Zhang L, Wang T, Valle D (2015) Reduced PLP2 expression increases ER-stress-induced neuronal apoptosis and risk for adverse neurological outcomes after hypoxia ischemia injury. Hum Mol Genet 24: 7221-7226.
  11. Shono Yuji, Yokota Chiaki, Kuge Yuji, Kido Shinsuke, Harada Akina, et al. (2011) Gene expression associated with an enriched environment after transient focal ischemia. Brain Res 1376: 60-65.
  12. Niu RN, Shang XP, Teng JF (2018) Overexpression of Egr2 and Egr4 protects rat brains against ischemic stroke by downregulating JNK signaling pathway. Biochimie 149: 62-70.
  13. Kalli AC, Reithmeier RAF (2018) Interaction of the human erythrocyte Band 3 anion exchanger 1 (AE1, SLC4A1) with lipids and glycophorin A: Molecular organization of the Wright (Wr) blood group antigen. PLoS Comput Biol 14: e1006284.
  14. Meng LL, Yuan SM, Tu CF, Liu G, L GX, et al. (2019) Next-generation sequencing identified a novel SPTB frameshift insertion causing hereditary spherocytosis in China. Ann Hematol 98: 223-226.
  15. Shao H, Im H, Castro CM, Breakefield X, Weissleder R, et al. (2018) New Technologies for Analysis of Extracellular Vesicles. Chem Rev 118: 1917-1950.
  16. Christensen M, Schratt GM (2009) microRNA involvement in developmental and functional aspects of the nervous system and in neurological diseases. Neurosci Lett 466: 55-62.
  17. Chen F, Du Y, Esposito E, Liu Y, Guo S, et al. (2015) Effects of Focal Cerebral Ischemia on Exosomal Versus Serum miR126. Transl Stroke Res 6: 478-484.
  18. Sorensen SS, Nygaard AB, Carlsen AL, Heegaard NHH, Bak M, et al. (2017) Elevation of brain-enriched miRNAs in cerebrospinal fluid of patients with acute ischemic stroke. Biomark Res 5: 24.
  19. Squadrito ML, Baer C, Burdet F, Maderna C, Gilfillan GD, et al. (2014) Endogenous RNAs modulate microRNA sorting to exosomes and transfer to acceptor cells. Cell Rep 8: 1432-1446.
  20. Pegtel DM, Peferoen L, Amor S (2014) Extracellular vesicles as modulators of cell-to-cell communication in the healthy and diseased brain. Philos Trans R Soc Lond B Biol Sci 369: 20130516.
  21. Kouhkan F, Hafizi M, Mobarra N, Mossahebi-Mohammadi M, Mohammadi S (2014) miRNAs: a new method for erythroid differentiation of hematopoietic stem cells without the presence of growth factors. Appl Biochem Biotechnol 172: 2055-2069.
  22. Listowski MA, Heger E, Boguslawska DM, Machnicka B, Kuliczkowski K, et al. (2013) microRNAs: fine tuning of erythropoiesis. Cell Mol Biol Lett 18: 34-46.
  23. Scherrer N, Fays F, Mueller B, Luft A, Fluri F, et al. (2017) MicroRNA 150-5p Improves Risk Classification for Mortality within 90 Days after Acute Ischemic Stroke. J Stroke 19: 323-332.
  24. He QW, Li Q, Jin HJ, Zhi F, Suraj B, et al. (2016) MiR-150 Regulates Poststroke Cerebral Angiogenesis via Vascular Endothelial Growth Factor in Rats. CNS Neurosci Ther 22: 507-517.
  25. Fang Z, He QW, Li Q, Chen XL, Baral S, et al. (2016) MicroRNA-150 regulates blood-brain barrier permeability via Tie-2 after permanent middle cerebral artery occlusion in rats. FASEB J 30: 2097-2107.
  26. Xiang M, Zeng Y, Yang R (2014) U6 is not a suitable endogenous control for the quantification of circulating microRNAs. Biochem Biophys Res Commun. 454: 210-214.
  27. Lawrie CH, Gal S, Dunlop HM (2008) Detection of elevated levels of tumour-associated microRNAs in serum of patients with diffuse large B-cell lymphoma. Br J Haematol 141: 672-675.
  28. Qian J, Jiang B, Li M, Xu H, Chen Z, et al. (2013) Prognostic significance of microRNA-16 expression in human colorectal cancer. World J Surg 37: 2944-2949. 
  29. Chen X, Liang H, Guan D, Wang C, Hu X, et al. (2013) A combination of Let-7d, Let-7g and Let-7i serves as a stable reference for normalization of serum microRNAs. PLoS One 8: e79652.