Long non-coding RNA MEG3 inhibits neovascularization in diabetic retinopathy by regulating microRNA miR-6720-5p and cytochrome B5 reductase 2

Figures & data

Table 1. Comparisons of baseline characteristic of subjects among the DR, NDR and control groups

Indicators	Control	NDR	DR
Gender (male/female)	17/13	18/12	16/14
Age (years)	46.39 ± 7.83	45.73 ± 8.33	48.61 ± 7.19
Course of disease (years)	/	8.15 ± 2.46	8.47 ± 2.92
BMI (kg/m^2)	23.96 ± 1.36	22.71 ± 1.82	23.41 ± 1.09
FBG (mmol/L)	4.04 ± 0.79	6.94 ± 2.15*	12.06 ± 2.27^*#
HbA1c (%)	4.12 ± 0.58	7.82 ± 2.24*	12.73 ± 1.47^*#
TC (mmol/L)	3.52 ± 0.37	6.04 ± 0.82*	4.38 ± 1.51*
HDL-C (mmol/L)	1.56 ± 0.43	1.35 ± 0.57	1.72 ± 0.64*
LDL-C (mmol/L)	1.89 ± 0.74	2.12 ± 1.33*	2.03 ± 1.25*
FPG (mmol/L)	5.27 ± 1.97	9.61 ± 2.46*	10.46 ± 2.32^*#
TG (mmol/L)	1.38 ± 0.41	4.41 ± 0.98*	4.63 ± 1.15*
BUN (mmol/L)	4.62 ± 0.73	7.94 ± 0.53*	8.02 ± 0.49*
FIns (pmol/mL)	3.04 ± 1.71	4.78 ± 2.56*	4.99 ± 1.04*
Cr (mmol/d)	52.77 ± 7.64	68.83 ± 1.76*	69.32 ± 1.37*
IL-1 (ng/L)	22.92 ± 7.15	29.55 ± 6.78*	35.22 ± 7.94^*#
IL-6 (ng/L)	95.74 ± 20.64	201.82 ± 27.69*	227.19 ± 46.26^*#
TNF-α (ng/L)	7.83 ± 1.96	20.15 ± 3.38*	20.97 ± 3.04*
VEGF (ng/L)	78.65 ± 7.38	93.57 ± 9.45*	109.25 ± 13.66^*#

Note: DR patients; NDR, non-DR diabetic patients; DR, diabetic retinopathy; FBG, fasting blood glucose; FPG, fasting plasma glucose; TG, triglyceride; TC, total cholesterol; HbA1c, glycosylated hemoglobin; BUN, blood urea nitrogen; Cr, creatinine; HDL-C, high-density lipoprotein cholesterol; LDL-C low-density lipoprotein cholesterol; FIns, fasting insulin; IL, interleukin; TNF, tumor necrosis factor; VEGF, vascular endothelial growth factor.

Table 2. Sequence of the primers used in this study

gene	primer type	Sequence
ACSM3	Forward	5’-AGGAAGATGCTACGTCATGCC-3’
	Reverse	5’-ATCCCCAGTTTGAAGTCCTGT-3’
DSCC1	Forward	5’-CGTGGTGATAAAGACGAGCA-3’
	Reverse	5’-CCGGAGTTTTACAACCAGGA-3’
SARDH	Forward	5’-AGCGACCTGACTGTTGCC-3’
	Reverse	5’-CCTGTAGCACCGTGTTTATG-3’
CYB5R2	Forward	5’-CCTTGTAGGGACCCGTCCC-3’
	Reverse	5’-GACAGGGGTGTAAGCCCTG-3’
EHHADH	Forward	5’-AAACTCAGACCCGGTTGAAGA-3’
	Reverse	5’-TTGCAGAGTCTACGGGATTCT − 3’
miR-494-3p	Forward	5’-AACGAGACGACGACAGAC-3’
	Reverse	5’-TGAAACATACACGGGAAACCTC-3’
MEG3	Forward	5’-CTCCCCTTCTAGCGCTCACG-3’
	Reverse	5’-CTAGCCGCCGTCTATACTACCGGCT-3’
GAPDH	Forward	5’-CAATGACCCCTTCATTGACC-3
	Reverse	5’-GACAAGCTTCCCGTTCTCAG-3’
U6	Forward	5’-AACGAGACGACGACAGAC-3
	Reverse	5’-GCAAATTCGTGAAGCGTTCCATA-3’

Figure 1. CYB5R2 and miR-6720-5p are associated with DR. (a) The DEGs in GSE53257 with adj.P < 0.05. GSE63257, mRNA expression profile including 6 DR samples and 5 non-DR samples. (b-g) The expression levels of ACSM3, DSCC1, SARDH, VISA, CYB5R2 and EHHADH in blood samples from DR, D and HC patients. DR, diabetic retinopathy. D, diabetes. HC, healthy control. (h) miR-494-3p, miR-6720-5p and miR-6512-3p were the common miRNAs binding to MEG3 and CYB5R2. starBase, an online tool for the prediction of miRNAs binding to MEG3. TargetScan, an online tool for the prediction of miRNAs binding to CYB5R2. (i-k) The expression levels of miR-494-3p, miR-6720-5p and miR-6512-3p in blood samples from DR, D and HC patients. DR, diabetic retinopathy. D, diabetes. HC, healthy control. *P < 0.05; **P < 0.001

Table 3. The downregulated DEGs in GSE53257

adj.P.Val	logFC	Gene.Symbol
0.000397	−3.08109	ACSM3
0.00386	−1.79671	DSCC1
0.010101	−1.38822	SARDH
0.010101	−1.31667	VISA
0.032456	−1.53842	CYB5R2
0.038058	−1.33973	EHHADH

Figure 2. Silencing MEG3 promotes neovascularization. (a) Serum levels of MEG3 in DR, D, HC groups. (b) The expression level of MEG3 in the retinal tissue microenvironment of DR and normal rat. (c)The expression level of MEG3 in the acute fluctuation of blood glucose level. (d) The expression level of MEG3 was measured by qRT-PCR in hRMECs treated with HG, NG or OS. (e) The expression level of MEG3 was detected by qRT-PCR in hRMECs transfected with si-MEG3. (f) Cell viability was detected by using MTT assay in hRMECs transfected with si-MEG3. (g) Cell apoptosis rate was detected by using flow cytometry assay in hRMECs transfected with si-MEG3. (h) Cell migration was assessed using transwell assay after treatment with si-MEG3. Representative images were shown. (i) Tube formation assay was used to detect tubulogenesis of hRMECs transfected with si-MEG3. Representative images were shown. DR, diabetic retinopathy. D, diabetes. HC, healthy control. HG, high glucose. OS, osmotic control. NG, normal glucose. *P < 0.05; **P < 0.001

Table 4. Pearson correlation analysis between plasma MEG3 and various indexes in the DR group

Indicators	MGE3 expression
	r	P
Age (years)	0.0231	0.724
Course of disease (years)	−0.217	0.009
BMI (kg/m^2)	−0.106	0.612
FBG (mmol/L)	0.001	0.319
HbA1c (%)	−0.584	<0.001
TC (mmol/L)	−0.019	0.943
HDL-C (mmol/L)	0.004	0.481
LDL-C (mmol/L)	−0.149	0.216
FPG (mmol/L)	−0.448	<0.001
TG (mmol/L)	−0.083	0.257
BUN (mmol/L)	−0.012	0.695
FIns (pmol/mL)	−0.057	0.472
Cr (mmol/d)	0.009	0.814
IL-1 (ng/L)	−0.196	0.271
IL-6 (ng/L)	0.045	0.793
TNF-α (ng/L)	−0.295	0.005
VEGF (ng/L)	−0.412	0.011

Figure 3. MEG3 acts on neovascularization by regulating miR-6720-5p. (a) Alignment of potential miR-6720-5p binding sites in MEG3. (b) Luciferase activity was detected in hRMECs transfected with constructs containing wild-type of MEG3 or mutated MEG3 plasmid in response to the transfection of miR-6720-5p mimic. (c) The expression level of miR-6720-5p in the retinal tissue microenvironment of DR and normal rat. (d) The expression level of miR-6720-5p was measured by qRT-PCR in hRMECs treated with HG, NG or OS. (e) The expression level of miR-6720-5p was detected by qRT-PCR in hRMECs transfected with si-MEG3 or miR-6720-5p inhibitor. (f) Cell viability was detected by using MTT assay in hRMECs transfected with si-MEG3 or miR-6720-5p inhibitor. (g) Cell apoptosis rate was detected by using flow cytometry assay in hRMECs transfected with si-MEG3 or miR-6720-5p inhibitor. (h) Cell migration was assessed using transwell assay after treatment with si-MEG3 or miR-6720-5p inhibitor. Representative images were shown. (i) Tube formation assay was used to detect tubulogenesis of hRMECs transfected with si-MEG3 or miR-6720-5p inhibitor. Representative images were shown. HG, high glucose. OS, osmotic control. NG, normal glucose. *P < 0.05; **P < 0.001

Figure 4. CYB5R2 is the target gene of miR-6720-5p, and knockdown of CYB5R2 reversed the inhibitory effect of miR-6720-5p inhibitor on neovascularization. (a) Alignment of potential CYB5R2 binding sites in miR-6720-5p. (b) Luciferase activity was detected in hRMECs transfected with constructs containing wild-type of CYB5R2 or mutated CYB5R2 plasmid in response to the transfection of miR-6720-5p mimic. (c) The expression level of CYB5R2 in the retinal tissue microenvironment of DR and normal rat. (d) The expression level of CYB5R2 was measured by qRT-PCR in hRMECs treated with HG, NG or OS. (e) The expression level of CYB5R2 was detected by qRT-PCR in hRMECs transfected with si-CYB5R2 or miR-6720-5p inhibitor. (f) Cell viability was detected by using MTT assay in hRMECs transfected with si-CYB5R2 or miR-6720-5p inhibitor. (g) Cell apoptosis rate was detected by using flow cytometry assay in hRMECs transfected with si-CYB5R2 or miR-6720-5p inhibitor. (h) Cell migration was assessed using transwell assay after treatment with si-CYB5R2 or miR-6720-5p inhibitor. Representative images were shown. (i) Tube formation assay was used to detect tubulogenesis of hRMECs transfected with si-CYB5R2 or miR-6720-5p inhibitor. Representative images were shown. HG, high glucose. OS, osmotic control. NG, normal glucose. *P < 0.05; **P < 0.001

Data availability statement

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.