The methylation status of OPG, RANK and RANKL genes in steroid-related ONFH patients and controls
The results of the methylation levels of OPG, RANK and RANKL genes in steroid-related ONFH patients and the control subjects are shown in Table 3. We found a CpG island in the OPG (chr8: 118951508 ~ 118952139) gene, in which 4 CpG units (OPG1_CpG_2, OPG2_CpG_1, OPG2_CpG_5 and OPG2_CpG_10.11) had a significantly higher methylation level than the control group. There are two CpG islands in the RANK (chr18: 62324637 ~ 62326523) gene. Among them, the methylation level of 15 CpG units (RANK1_CpG_6, RANK1_CpG_15, RANK1_CpG_20, RANK1_CpG_21, RANK2_CpG_10.11, RANK2_CpG_15, RANK2_CpG_16, RANK2_CpG_21.22, RANK2_CpG_23.24.25.26.27, RANK2_CpG_29, RANK2_CpG_30.31, RANK2_CpG_36, RANK3_CpG_1, RANK3_CpG_8 and RANK3_CpG_10) is higher than that of the control group, and the difference is statistically significant. And there is a CpG island in the RANKL (chr13: 42573413 ~ 42574559) gene, the methylation level of 10 CpG units (RANKL1_CpG_1, RANKL1_CpG_11, RANKL1_CpG_12.13, RANKL1_CpG_19, RANKL1_CpG_21, RANKL1_CpG_24, RANKL1_CpG_25.26.27, RANKL1_CpG_44.45, RANKL2_CpG_2 and RANKL2_CpG_9.10.11.12) in the control group was significantly lower than that of steroid-related ONFH. Through the histogram comparison (Fig. 2), we can clearly see the CpG units that are significantly different between the two groups.
Table 3
Methylation levels of OPG, RANK and RANKL genes in the case group and control group.
Gene_ID
|
Number of subjects
|
Methylation levels %
|
p value
|
Control
|
Case
|
Control (mean ± SD)
|
Case (mean ± SD)
|
OPG1_CpG_2
|
49
|
50
|
0.06 ± 0.01
|
0.07 ± 0.01
|
0.046 a
|
OPG2_CpG_1
|
50
|
49
|
0.08 ± 0.09
|
0.13 ± 0.10
|
0.016 b
|
OPG2_CpG_5
|
47
|
47
|
0.05 ± 0.04
|
0.07 ± 0.06
|
0.032 a
|
OPG2_CpG_10.11
|
48
|
50
|
0.04 ± 0.02
|
0.05 ± 0.03
|
0.020 b
|
RANK1_CpG_6
|
48
|
49
|
0.16 ± 0.04
|
0.09 ± 0.08
|
0.035 a
|
RANK1_CpG_15
|
49
|
50
|
0.82 ± 0.25
|
0.93 ± 0.20
|
0.001 a
|
RANK1_CpG_20
|
49
|
50
|
0.07 ± 0.09
|
0.12 ± 0.15
|
0.005 a
|
RANK1_CpG_21
|
48
|
47
|
0.13 ± 0.17
|
0.23 ± 0.22
|
0.044 a
|
RANK2_CpG_10.11
|
46
|
50
|
0.07 ± 0.04
|
0.17 ± 0.19
|
0.000 a
|
RANK2_CpG_15
|
47
|
50
|
0.04 ± 0.04
|
0.09 ± 0.08
|
0.001 a
|
RANK2_CpG_16
|
46
|
48
|
0.05 ± 0.03
|
0.08 ± 0.02
|
0.000 a
|
RANK2_CpG_21.22
|
47
|
50
|
0.02 ± 0.01
|
0.03 ± 0.03
|
0.043 a
|
RANK2_CpG_23.24.25.26.27
|
47
|
50
|
0.04 ± 0.02
|
0.07 ± 0.03
|
0.000 a
|
RANK2_CpG_29
|
47
|
50
|
0.01 ± 0.02
|
0.03 ± 0.04
|
0.007 a
|
RANK2_CpG_30.31
|
45
|
49
|
0.04 ± 0.02
|
0.05 ± 0.01
|
0.000 a
|
RANK2_CpG_36
|
47
|
50
|
0.03 ± 0.03
|
0.04 ± 0.03
|
0.019 a
|
RANK3_CpG_1
|
46
|
50
|
0.38 ± 0.09
|
0.41 ± 0.07
|
0.023 a
|
RANK3_CpG_8
|
46
|
50
|
0.05 ± 0.03
|
0.09 ± 0.00
|
0.000 b
|
RANK3_CpG_10
|
47
|
50
|
0.03 ± 0.02
|
0.04 ± 0.01
|
0.007 a
|
RANKL1_CpG_1
|
43
|
49
|
0.07 ± 0.05
|
0.12 ± 0.11
|
0.009 b
|
RANKL1_CpG_11
|
45
|
50
|
0.02 ± 0.03
|
0.03 ± 0.02
|
0.014 a
|
RANKL1_CpG_12.13
|
44
|
49
|
0.10 ± 0.07
|
0.11 ± 0.03
|
0.004 a
|
RANKL1_CpG_19
|
43
|
49
|
0.07 ± 0.15
|
0.10 ± 0.14
|
0.001 a
|
RANKL1_CpG_21
|
44
|
50
|
0.06 ± 0.08
|
0.08 ± 0.14
|
0.041 a
|
RANKL1_CpG_24
|
46
|
50
|
0.11 ± 0.15
|
0.13 ± 0.16
|
0.018 a
|
RANKL1_CpG_25.26.27
|
45
|
50
|
0.30 ± 0.19
|
0.35 ± 0.15
|
0.000 a
|
RANKL1_CpG_44.45
|
45
|
50
|
0.03 ± 0.01
|
0.04 ± 0.01
|
0.000 b
|
RANKL2_CpG_2
|
48
|
50
|
0.10 ± 0.06
|
0.15 ± 0.09
|
0.016 b
|
RANKL2_CpG_9.10.11.12
|
48
|
50
|
0.05 ± 0.04
|
0.07 ± 0.02
|
0.000 a
|
p ≤ 0.05 indicates statistical significance.
a Mann-Whitney U test
b Independent samples t test.
|
The differences in the methylation rates of OPG, RANK and RANKL genes between steroid-related ONFH and control subjects are listed in Table 4. Compared with control subjects, the methylation rate of 3 CpG units (OPG2_CpG_1, OPG2_CpG_5 and OPG2_CpG_10.11) in the OPG gene, 10 CpG units (RANK1_CpG_2, RANK1_CpG_6, RANK1_CpG_18, RANK1_CpG_20, RANK1_CpG_24.25.26.27, RANK2_CpG_10.11, RANK2_CpG_15, RANK2_CpG_23.24.25.26.27, RANK2_CpG_36 and RANK3_CpG_8) in the RANK gene and 6 CpG units (RANKL1_CpG_14.15, RANKL1_CpG_19, RANKL1_CpG_21, RANKL1_CpG_24, RANKL2_CpG_2 and RANKL2_CpG_9.10.11.12) in the RANKL gene of ONFH patients was significantly increased.
Table 4
Relationship between methylation rate and steroid-related osteonecrosis of femoral head.
Gene_ID
|
Group
|
Number of subjects
|
p value *
|
Non-Meth.
|
Meth.
|
OPG2_CpG_1
|
Control
|
27
|
23
|
0.001
|
Case
|
10
|
39
|
OPG2_CpG_5
|
Control
|
30
|
17
|
0.039
|
Case
|
20
|
27
|
OPG2_CpG_10.11
|
Control
|
42
|
6
|
0.007
|
Case
|
32
|
18
|
RANK1_CpG_2
|
Control
|
17
|
32
|
0.008
|
Case
|
6
|
44
|
RANK1_CpG_6
|
Control
|
18
|
30
|
0.004
|
Case
|
6
|
43
|
RANK1_CpG_18
|
Control
|
16
|
33
|
0.028
|
Case
|
7
|
43
|
RANK1_CpG_20
|
Control
|
34
|
15
|
0.001
|
Case
|
18
|
32
|
RANK1_CpG_24.25.26.27
|
Control
|
16
|
32
|
0.013
|
Case
|
6
|
43
|
RANK2_CpG_10.11
|
Control
|
19
|
27
|
0.012
|
Case
|
9
|
41
|
RANK2_CpG_15
|
Control
|
33
|
14
|
0.003
|
Case
|
20
|
30
|
RANK2_CpG_23.24.25.26.27
|
Control
|
33
|
14
|
0.005
|
Case
|
21
|
29
|
RANK2_CpG_36
|
Control
|
40
|
7
|
0.048
|
Case
|
34
|
16
|
RANK3_CpG_8
|
Control
|
25
|
21
|
0.000
|
Case
|
10
|
40
|
RANKL1_CpG_14.15
|
Control
|
16
|
28
|
0.044
|
Case
|
9
|
41
|
RANKL1_CpG_19
|
Control
|
23
|
20
|
0.015
|
Case
|
14
|
35
|
RANKL1_CpG_21
|
Control
|
34
|
10
|
0.018
|
Case
|
27
|
23
|
RANKL1_CpG_24
|
Control
|
26
|
20
|
0.027
|
Case
|
17
|
33
|
RANKL2_CpG_2
|
Control
|
18
|
30
|
0.003
|
Case
|
6
|
44
|
RANKL2_CpG_9.10.11.12
|
Control
|
29
|
19
|
0.000
|
Case
|
9
|
41
|
Non-Meth Non-Methylation; Meth Methylation.
p < 0.05 indicates statistical significance
* chi-square test
|
Through the above comparative analysis, we found that the methylation status of OPG, RANK and RANKL genes is indeed different in steroid-induced ONFH patients and healthy controls. In order to further study the effect of methylation on steroid-induced ONFH, we divided the methylation levels into hypermethylation and hypomethylation according to the median of the control group, and performed logistic regression analysis. The results in Table 5 showed that, in the OPG gene, the high methylation level of 2 CpG sites (OPG2_CpG_1 and OPG2_CpG_5) increases the risk of steroid-induced ONFH. In the RANK gene, the increase in the methylation level of 12 CpG units (RANK1_CpG_6, RANK1_CpG_15, RANK1_CpG_20, RANK2_CpG_10.11, RANK2_CpG_15, RANK2_CpG_16, RANK2_CpG_21.22, RANK2_CpG_23.24.25.26.27, RANK2_CpG_29, RANK2_CpG_36, RANK3_CpG_8 and RANK3_CpG_10) is positively correlated with the risk of steroid-induced ONFH. In the RANKL gene, the increase in the methylation level of 10 CpG units (RANKL1_CpG_1, RANKL1_CpG_12.13, RANKL1_CpG_14.15, RANKL1_CpG_19, RANKL1_CpG_21, RANKL1_CpG_24, RANKL1_CpG_25.26.27, RANKL1_CpG_44.45, RANKL1_CpG_46 and RANKL2_CpG_9.10.11.12) also increases the risk of steroid-induced ONFH.
Table 5
Relationship between methylation levels and steroid-associated osteonecrosis of femoral head.
Gene_ID
|
OR
|
95% CI
|
p value
|
lower limit
|
upper limit
|
OPG2_CpG_1
|
4.62
|
1.95
|
10.95
|
0.001
|
OPG2_CpG_5
|
3.45
|
1.32
|
9.02
|
0.005
|
RANK1_CpG_6
|
4.83
|
1.64
|
14.25
|
0.004
|
RANK1_CpG_15
|
4.67
|
1.82
|
12.01
|
0.001
|
RANK1_CpG_20
|
4.25
|
1.80
|
10.05
|
0.001
|
RANK2_CpG_10.11
|
4.76
|
1.95
|
11.65
|
0.001
|
RANK2_CpG_15
|
4.17
|
1.69
|
10.29
|
0.002
|
RANK2_CpG_16
|
3.71
|
1.30
|
10.59
|
0.014
|
RANK2_CpG_21.22
|
2.90
|
1.19
|
7.05
|
0.019
|
RANK2_CpG_23.24.25.26.27
|
3.43
|
1.45
|
8.10
|
0.005
|
RANK2_CpG_29
|
3.48
|
1.36
|
8.91
|
0.009
|
RANK2_CpG_36
|
2.95
|
1.27
|
6.86
|
0.012
|
RANK3_CpG_8
|
6.56
|
2.54
|
16.93
|
0.000
|
RANK3_CpG_10
|
3.91
|
1.65
|
9.27
|
0.002
|
RANKL1_CpG_1
|
12.66
|
4.10
|
39.11
|
0.000
|
RANKL1_CpG_12.13
|
3.57
|
1.42
|
8.98
|
0.007
|
RANKL1_CpG_14.15
|
3.10
|
1.30
|
7.38
|
0.011
|
RANKL1_CpG_19
|
6.92
|
2.66
|
18.00
|
0.000
|
RANKL1_CpG_21
|
3.05
|
1.30
|
7.12
|
0.010
|
RANKL1_CpG_24
|
2.44
|
1.05
|
5.64
|
0.038
|
RANKL1_CpG_25.26.27
|
5.89
|
2.32
|
14.93
|
0.000
|
RANKL1_CpG_44.45
|
3.84
|
1.53
|
9.62
|
0.004
|
RANKL1_CpG_46
|
3.74
|
1.40
|
9.97
|
0.008
|
RANKL2_CpG_9.10.11.12
|
3.18
|
1.31
|
7.73
|
0.011
|
OR odds ratio; CI: confidence interval.
p < 0.05 indicates statistical significance
p values were calculated by logistic regression adjusted for age and gender.
|
Predictive indicators of steroid-related ONFH
In order to further study the applicability of OPG, RANK and RANKL gene methylation levels as potential diagnostic biomarkers for steroid-related ONFH, follow-up ROC analysis was performed on the above data (Table 6). In the OPG gene, the Area Under Curve (AUC) values of 3 CpG units (OPG2_CpG_1, OPG2_CpG_5 and OPG2_CpG_10.11) are greater than 0.5, of which OPG2_CpG_1 (AUC: 0.679 95% CI: 0.57–0.79 p = 0.002) has a higher diagnostic value (Fig. 3A); In the RANK gene, the AUC values of 15 CpG units (RANK1_CpG_6, RANK1_CpG_15, RANK1_CpG_20, RANK1_CpG_21, RANK2_CpG_10.11, RANK2_CpG_15, RANK2_CpG_16, RANK2_CpG_21.22, RANK2_CpG_23.24.25.26.27, RANK2_CpG_29, RANK2_CpG_30.31, RANK2_CpG_36, RANK3_CpG_1, RANK3_CpG_8 and RANK3_CpG_10) are greater than 0.5, of which RANK3_CpG_8 (AUC: 0.778 95% CI: 0.69–0.87 p = 0.000) has a higher diagnostic value (Fig. 3B); In the RANKL gene, 10 CpG units (RANKL1_CpG_1, RANKL1_CpG_11, RANKL1_CpG_12.13, RANKL1_CpG_19, RANKL1_CpG_21, RANKL1_CpG_24, RANKL1_CpG_25.26.27, RANKL1_CpG_44.45, RANKL2_CpG_2 and RANKL2_CpG_9.10.11.12) have AUC values greater than 0.5, of which RANKL1_CpG_1 (AUC: 0.764 95% CI: 0.66–0.87 p = 0.000) is the site with the most diagnostic value (Fig. 3C).
Table 6
Receiver operating characteristic curve analyses for predicting steroid-associated osteonecrosis of femoral head prevalence.
Gene_ID
|
Area
|
Sema
|
Youden's index
|
95% CI
|
p value b
|
lower limit
|
upper limit
|
OPG2_CpG_1
|
0.679
|
0.055
|
0.355
|
0.57
|
0.79
|
0.002
|
OPG2_CpG_5
|
0.628
|
0.059
|
0.319
|
0.51
|
0.74
|
0.032
|
OPG2_CpG_10.11
|
0.622
|
0.057
|
0.235
|
0.51
|
0.73
|
0.037
|
RANK1_CpG_6
|
0.624
|
0.057
|
0.253
|
0.51
|
0.74
|
0.036
|
RANK1_CpG_15
|
0.669
|
0.055
|
0.330
|
0.56
|
0.78
|
0.004
|
RANK1_CpG_20
|
0.662
|
0.055
|
0.334
|
0.56
|
0.77
|
0.005
|
RANK1_CpG_21
|
0.619
|
0.059
|
0.324
|
0.50
|
0.74
|
0.045
|
RANK2_CpG_10.11
|
0.750
|
0.051
|
0.450
|
0.65
|
0.85
|
0.000
|
RANK2_CpG_15
|
0.697
|
0.053
|
0.394
|
0.59
|
0.80
|
0.001
|
RANK2_CpG_16
|
0.760
|
0.050
|
0.442
|
0.66
|
0.86
|
0.000
|
RANK2_CpG_21.22
|
0.617
|
0.057
|
0.226
|
0.51
|
0.73
|
0.047
|
RANK2_CpG_23.24.25.26.27
|
0.718
|
0.051
|
0.352
|
0.62
|
0.82
|
0.000
|
RANK2_CpG_29
|
0.655
|
0.056
|
0.295
|
0.55
|
0.76
|
0.008
|
RANK2_CpG_30.31
|
0.688
|
0.055
|
0.301
|
0.58
|
0.80
|
0.002
|
RANK2_CpG_36
|
0.636
|
0.057
|
0.256
|
0.53
|
0.75
|
0.021
|
RANK3_CpG_1
|
0.635
|
0.058
|
0.315
|
0.52
|
0.75
|
0.023
|
RANK3_CpG_8
|
0.778
|
0.047
|
0.411
|
0.69
|
0.87
|
0.000
|
RANK3_CpG_10
|
0.699
|
0.053
|
0.318
|
0.59
|
0.80
|
0.001
|
RANKL1_CpG_1
|
0.764
|
0.054
|
0.479
|
0.66
|
0.87
|
0.000
|
RANKL1_CpG_11
|
0.645
|
0.057
|
0.256
|
0.53
|
0.76
|
0.015
|
RANKL1_CpG_12.13
|
0.672
|
0.059
|
0.327
|
0.58
|
0.79
|
0.004
|
RANKL1_CpG_19
|
0.697
|
0.056
|
0.420
|
0.59
|
0.81
|
0.001
|
RANKL1_CpG_21
|
0.622
|
0.059
|
0.271
|
0.51
|
0.74
|
0.042
|
RANKL1_CpG_24
|
0.640
|
0.058
|
0.302
|
0.53
|
0.75
|
0.019
|
RANKL1_CpG_25.26.27
|
0.710
|
0.056
|
0.411
|
0.60
|
0.82
|
0.000
|
RANKL1_CpG_44.45
|
0.734
|
0.052
|
0.396
|
0.63
|
0.84
|
0.000
|
RANKL2_CpG_2
|
0.629
|
0.057
|
0.278
|
0.52
|
0.74
|
0.027
|
RANKL2_CpG_9.10.11.12
|
0.738
|
0.052
|
0.438
|
0.64
|
0.84
|
0.000
|
Sem Standard error of mean; CI Confidence index.
p < 0.05 indicates statistical significance
a Under non-parametric assumptions
b Null hypothesis: real area = 0.5
|